97/3CR
International Court Cour internationale
of Justice de Justice
THE HAGUE LA HAYE
YEAR 1997
Public sitting
held on Tuesday 4 March 1997, at 10 a.m., at the Peace Palace,
President Schwebel presiding
in the case concerning Gab_íkovo-Nagymaros Project
(Hungary/Slovakia)
_______________
VERBATIM RECORD
_______________
ANNEE 1997
Audience publique
tenue le mardi 4 mars 1997, à 10 heures, au Palais de la Paix,
sous la présidence de M. Schwebel, Président
en l'affaire relative au Projet Gab_íkovo-Nagymaros
(Hongrie/Slovaquie)
____________
COMPTE RENDU
____________
Present:chwesilent -2-
Vieceeranirynt
Oda Judges
Bedjaoui
Guillaume
Ranjeva
Herczegh
Shi
Fleischhauer
Koroma
Vereshchetin
Parra-Aranguren
Kooijmans
Rezek
Judge ad hoc Skubiszewski
Valencia-ORsenistrar -3-
Présents : M. Schwebel, Président
WVier-mrs.re,nt
Oda MM.
Bedjaoui
Guillaume
Ranjeva
Herczegh
Shi
Fleischhauer
Koroma
Vereshchetin
Parra-Aranguren,
Kooijmans
juges Rezek,
Sugbeiszewski, ad hoc
ValGnrcaf.erspina, -4-
The Republic of Hungary is represented by:
H.E. Mr. György Szénási, Ambassador, Head of the International Law
Department, Ministry of Foreign Affairs,
as Agent and Counsel;
H.E. Mr. Dénes Tomaj, Ambassador of the Republic of Hungary to the
Netherlands,
Coa-Agent;
Mr. James Crawford, Whewell Professor of International Law,
University of Cambridge,
Mr. Pierre-Marie Dupuy, Professor at the University Panthéon-Assas
(Paris II) and Director of the Institut des hautes études
internationales of Paris,
Mr. Alexandre Kiss, Director of Research, Centre National de la
recherche Scientifique (ret.),
Mr. László Valki, Professor of International Law, Eötvös Lorand
University, Budapest,
Mr. Boldizsár Nagy, Associate Professor of International Law,
Eötvös Loránd University, Budapest,
Mr. Philippe Sands, Reader in International Law, University of
London, School of Oriental and African Studies, and Global
Professor of Law, New York University,
Ms Katherine Gorove, consulting Attorney,
as Counsel and Advocates;
Dr. Howard Wheater, Professor of Hydrology, Imperial College,
London,
Dr. Gábor Vida, Professor of Biology, Eötvös Loránd University,
Budapest, Member of the Hungarian Academy of Sciences,
Dr. Roland Carbiener, Professor emeritus of the University of
Strasbourg,
Dr. Klaus Kern, consulting Engineer, Karlsruhe,
Adavocates;
Mr. Edward Helgeson,
Mr. Stuart Oldham,
Adavisers; -5-
La République de Hongrie est représentée par :
S. Exc. M. György Szénási, ambassadeur, directeur du département du
droit international au ministère des affaires étrangères,
comme agent et conseil;
S. Exc. M. Dénes Tomaj, ambassadeur de la République de Hongrie aux
Pays-Bas,
comcoagent ;
M. James R. Crawford, professeur de droit international, titulaire
de la chaire Whewell à l'Université de Cambridge,
M. Pierre-Marie Dupuy, professeur à l'Université Panthéon-Assas
(Paris II) et directeur de l'Institut des hautes études
internationales de Paris,
M. Alexandre Kiss, directeur de recherches au Centre national de la
recherche scientifique (en retraite),
M. Lászlo Valki, professeur de droit international à l'Université
Eötvös Lorand de Budapest,
M. Boldizsár Nagy, professeur associé de droit international à
l'Université Eötvös Lorand de Budapest,
M. Philippe Sands, chargé de cours en droit international à
l'Université de Londres, School of Oriental and African Studies,
et Global Professor of Law à l'Université de New York,
Mme Katherine Gorove, juriste-conseil,
comme conseils et avocats;
M. Howard Wheater, professeur d'hydrologie à l'Imperial College de
Londres,
M. Gábor Vida, professeur de biologie à l'Université Eötvös Lorand
de Budapest, membre de l'Académie des sciences de Hongrie,
M. Roland Carbiener, professeur émérite de l'Université de
Strasbourg,
M. Klaus Kern, ingénieur-conseil à Karlsruhe,
comavocats ;
M. Edward Helgeson,
M. Stuart Oldham,
comconseillers ; -6-
Dr. György Kovács,
Mr. Timothy Walsh,
as Technical Advisers;
Dr. Attila Nyikos,
Assisstant;
Ms Éva Kocsis,
Ms Katinka Tompa,
Searsetaries.
The Republic of Slovakia is represented by:
H.E. Dr. Peter Tomka, Ambassador, Legal Adviser of the Ministry of
Foreign Affairs,
Agaent;
Dr. Václav Mikulka, Member of the International Law Commission,
as Co-Agent, Counsel and Advocate;
Mr. Derek W. Bowett, C.B.E., Q.C., F.B.A., Emeritus Whewell
Professor of International Law at the University of Cambridge,
Former Member of the International Law Commission,
Coausnsel;
Mr. Stephen C. McCaffrey, Professor of International Law at the
University of the Pacific, McGeorge School of Law, Sacramento,
United States of America, Former Member of the International Law
Commission,
Mr. Alain Pellet, Professor at the University of Paris X — Nanterre
and at the Institute of Political Studies, Paris, Member of the
International Law Commission,
Mr. W. Walter D. Sohier, Member of the Bar of the State of New York
and of the District of Colombia,
Sir Arthur Watts, K.C.M.G., Q.C., Barrister, Member of the Bar of
England and Wales,
Mr. Samuel S. Wordsworth, avocat à la Cour au barreau de Paris,
Solicitor England and Wales, Frere Cholmeley, Paris,
as Counsel and Advocates;
Mr. Igor Mucha, Professor of Hydrogeology and Former Head of the
Groundwater Department at the Faculty of Natural Sciences of
Comenius University in Bratislava, -7-
M. György Kovács,
M. Timothy Walsh,
comme conseillers techniques;
M. Attila Nyikos,
comassistant ;
Mme Éva Kocsis,
Mme Katinka Tompa,
comsecreétaires .
La République slovaque est representée par :
S. Exc. M. Peter Tomka, ambassadeur, conseiller juridique du
ministère des affaires étrangères,
comament ;
M. Václav Mikulka, membre de la Commission du droit international,
comme coagent, conseil et avocat;
M. Derek W. Bowett, C.B.E., Q.C., F.B.A., professeur émérite, ancien
titulaire de la chaire Whewell à l'Université de Cambridge, ancien
membre de la Commission du droit international,
comconseil ;
M. Stephen C. McCaffrey, professeur de droit international à la
faculté de droit McGeorge de l'Université du Pacifique, Sacramento
(Etats-Unis d'Amérique), ancien membre de la Commission du droit
international,
M. Alain Pellet, professeur à l'Université de Paris X-Nanterre et à
l'Institut d'études politiques de Paris, membre de la Commission du droit
international,
M. Walter D. Sohier, membre des barreaux de l'Etat de New York et du
district de Columbia,
Sir Arthur Watts, K.C.M.G., Q.C., avocat au barreau d'Angleterre et
du pays de Galles,
M. Samuel S. Wordsworth, avocat à la Cour, Frere Cholmeley, Paris,
Solicitor auprès de la Cour suprême d'Angleterre et du pays de
Galles,
comme conseils et avocats;
M. Igor Mucha, professeur d'hydrogéologie et ancien directeur du
département des eaux souterraines à la faculté des sciences
naturelles de l'Université Comenius de Bratislava,
Mr. Karra Venkateswara Rao, Director of Water Resources Engineering, -8-
Department of Civil Engineering, City University, London,
Mr. Jens Christian Refsgaard, Head of Research and Development,
Danish Hydraulic Institute,
as Counsel and Experts;
Dr. Cecília Kandrá_ová, Director of Department, Ministry of Foreign
Affairs,
Mr. Lud_k Krajhanzl, Attorney at Law, Vyroubal Krajhanzl Skácel and
Partners Law Firm, Prague,
Mr. Miroslav Liška, Head of the Division for Public Relations and
Expertise, Water Resources Development State Enterprise,
Bratislava,
Dr. Peter Vršanský, Minister-Counsellor, chargé d'affaires a.i. of
the Embassy of the Slovak Republic, The Hague,
Coausnsellors;
Ms Anouche Beaudouin, allocataire de recherche at the University
of Paris X — Nanterre,
Ms Cheryl Dunn, Frere Cholmeley, Paris,
Ms Nikoleta Glindová, attachée, Ministry of Foreign Affairs,
Mr. Drahoslav Štefánek, attaché, Ministry of Foreign Affairs,
as Legal Assistants. -9-
M. Karra Venkateswara Rao, directeur du Génie, section des
ressources hydrologiques, département du Génie civil, Université
de la ville de Londres,
M. Jens Christian Refsgaard, directeur de la recherche et du
développement à l'Institut danois d'hydraulique,
comme conseils et experts;
Mme Cecília Kandrá_ová, directeur de département, ministère des
affaires étrangères,
M. Lud_k Krajhanzl, avocat, membre du cabinet Vyroubal Krajhanzl
Skácel et associés, Prague,
M. Miroslav Liška, directeur de la division des relations publiques
et de l'expertise, entreprise d'Etat pour le développement des
ressources hydrauliques, Bratislava,
M. Peter Vršanský, ministre-conseiller, chargé d'affaires a.i. à
l'ambassade de la République slovaque, La Haye,
cocomnseillers ;
Mlle Anouche Beaudouin, allocataire de recherche à l'Université de
Paris X-Nanterre,
Mme Cheryl Dunn, Frere Cholmeley, Paris,
Mme Nikoleta Glindová, attachée, ministère des affaires étrangères,
M. Drahoslav Štefánek, attaché, ministère des affaires étrangères,
comme assistants juridiques. -10-
The PRESIDENT: Please be seated. We continue this morning with the oral presentations of
the Republic of Hungary and I should like first to call upon Professor Nagy.
Mr. NAGY:
6. THE OBJECT AND PURPOSE OF THE TREATY AND THE
CHARACTER OF THE ORIGINAL PROJECT
I. Introduction
1. Mr President, Members of the Court. It is an honour to appear before you for the first
time. My task this morning is to clarify disputed issues concerning the purposes and object of the
1977 Treaty and the way how the regulatory regime of the Project functioned.
2. The 1977 Treaty had as its purpose the creation of a politically motivated economic joint
venture which was to be consistent with environmental protection. In this presentation I will
address the object and purpose of the Treaty, its character as a framework treaty to be flexibly
applied, and its connection to various related agreements, as well as to the Joint Contractual Plan.
3. The picture I will describe is of a Treaty having two main purposes: first, the
strengthening of fraternal relations between two States and contributing to the socialist integration
among COMECON States, and second, utilising the natural resources of the Bratislava-Budapest
section of the Danube river for the development of energy and, to a lesser and incidental extent,
other sectors. These are the purposes identified in the Treaty's Preamble. The Barrage System was
to be jointly constructed and jointly operated. The Treaty did not establish a territorial regime. It
did not alienate either party’s rights over its share of the water. And it was intended to function
without environmental degradation and in an economically viable manner.
4. I will address in turn the following aspects: first, the Treaty's objects and purposes;
second, its character as a joint investment; third, its relationship with other agreements; and fourth,
its character as an instrument to be flexibly applied. I begin with the first point.
II. The Purposes of the 1977 Treaty
5. Properly characterizing the purpose of the 1977 Treaty is a significant issue, and one on
which the Parties are not in agreement. Arguments concerning material breach, supervening
impossibility of performance, fundamental change of circumstances and even "approximate
application" depend upon the answers to the question: what was the object and purpose of the
1977 Treaty and its related agreements? -11-
6. For Slovakia, the basic aims of the Treaty were such that even through Variant C they
"could still be achieved, at least insofar as the Bratislava-Sap (Palkovicovo) stretch was
concerned" . The extent to which this is true depends upon whether or not the Treaty incorporates
the idea of two riparian States cooperating in the realization of an agreed joint venture. If it does,
as Hungary believes, then Variant C is plainly incompatible with the 1977 Treaty.
7. For its part Hungary has consistently maintained that the 1977 Treaty had two purposes:
the political and the economic . Slovakia by contrast identifies almost a dozen items under
various headings. The chart you see on the screen, which is also contained in your folders,
illustrates the proliferation of Slovak views as to the Treaty's basic objectives (Illus. 6.1). We are
told, for example, that the suspended works were "the very object of the Treaty itself" . Elsewhere
it is said that the Treaty sought "the improvement of surface and ground water" , or "environment
18
protection" . And yet elsewhere the "monitoring system" is elevated to the status of a "basic
19
aim" .
8. We certainly agree with Slovakia that the objects and purposes of the Treaty may be
ascertained not only from the treaty but from its negotiating history . As Professor Valki has
described, that history took place at a time, from the late 1950's to 1977, which has been
characterized by Slovak Prime Minister Carnogursky as a totalitarian era. Writing to Hungarian
Prime Minister Antall in 1991 a week after his Government's resolution approving the details of
Variant C, he says: "Both of our countries are now living in a period of transition from
21
totalitarianism to democratic systems." The Project was a plan conceived by two totalitarian
14SM, para. 5.26.
15
HM, paras. 4.04-4.06, HCM, para. 1.13, HR paras. 1.03-1.11.
16
SR, para. 2.12.
17SM, para. 6.132.
18SCM, para. 10.125.
19
SM, para. 5.26.
20SM, para. 8.73.
21SM, Vol. IV, Annex 93, HM, Vol. 4, Annex 56. -12-
States and partly realized by them. "Comrade Vohsalik's report from 1955 annexed by Slovakia 22
illustrates this point. He describes a meeting with Malenkov of the Central Committee of the
Soviet Communist Party, at which the former Soviet Prime Minister described the construction of
23
the water works on the Danube as being 'mainly a political issue' since it concerned other
Danubian countries. Malenkov stressed that the discussions on an agreement between
Czechoslovakia and Hungary had to be 'held on a high political level'." 24 When he also expressed
the Soviet Union's readiness to participate in these discussions, his interlocutor knew that this was
an offer having the force of a command. And so it was.
9. The political character of the 1977 Treaty is reflected in the fact that the scheme emerged
from consultations with Soviet participation between 1956 and 1960, and that it was approved by
the COMECON Commission for Energy and Agriculture in September 1961 . The documents
annexed by Slovakia confirm the Hungarian argument 26 as to the political character of the 1977
Treaty and the Project.
10. Hungary and Slovakia also agree that the Project was intended to have certain economic
benefits in the fields of energy production, flood protection, and navigation. They differ on their
assessment of the magnitude of these benefits and in the evaluation of viable alternatives. Later
today Ms Gorove will deal with the viability of the Project so far as issues of energy production,
flood protection, navigation and other sectors are concerned.
11. No doubt the Project would have brought benefits and imposed costs in several sectors.
However, in identifying the object and purposes of the 1977 Treaty, whether for the purpose of
treaty modification, termination or "approximate application", the reasonable approach is to remain
with the intention of the Parties as expressed in the Treaty itself, as well as its history. It is not
appropriate to pick and choose among the elements listed in the preamble of the 1977 Treaty,
elevating some of them to the "main objective" and leaving others in the shadow, as Slovakia does.
22
SCM, Vol. 2, Annex 2.
23
SCM, Vol. 2, Annex 2.
24Ibid.
25
SM, Vol. II, Annex 3, at p. 35.
26See especially HR, Vol. 2, App. 3 and HM, paras. 3.12-3.28. -13-
And as I have said, from the Treaty it is clear that the Parties had two purposes in mind: the
political and the economic, which they sought to realize through the "joint investment".
12. But these purposes and the "joint investment" had to be consistent with environmental
protection. This is clearly reflected in the fact that the 1977 Treaty devoted no less than three
articles to the protection of water, nature, and fisheries. In that sense it might be said that the
Treaty accurately sought to prepare the Parties for the environmental changes which were sure to
come, and to which both Parties had committed themselves in the 1975 Helsinki Final Act of the
Conference on Security and Cooperation in Europe, of which Ms Gorove will say more later this
morning. By now you will no doubt be familiar with those three Articles, the requirements of
which were mandatory. Article 15 of the Treaty provided that the parties "shall ensure ... that the
quality of the water in the Danube is not impaired as a result of the construction and operation" of
the Barrage System. Article 19 provided that the parties "shall ... ensure compliance with the
obligations for the protection of nature". And Article 20 provided that the parties "shall take
appropriate measures for the protection of fishing interests" in conformity with the 1958 Danube
Fisheries Agreement. In other words, the 1977 Treaty committed the parties to ensuring
conformity with applicable environmental norms from time to time.
III. The Character of the Project as a "Joint Investment"
13. I turn now to my second point, the character of the project as a "joint investment". The
object of the Treaty was the construction and operation of the Project consisting of the two
barrages under joint control. Its character was a large-scale industrial and infrastructural
investment reflecting the readiness of socialist States to allocate enormous financial resources in
order to achieve political and economic gains. But that investment had to conform to economic
and environmental standards throughout its lifetime. The Treaty did not establish a new,
immutable territorial régime regulating sovereign rights of States. It remained a socialist business
transaction in which not only companies, but also States participated, because of the all pervasive
role of the State in the economy during that period.
14. The Barrage System was to be a large-scale investment under socialist conditions. An
investment under socialist conditions means a venture of a predominantly economic nature but
lacking the discipline of market forces or of democratic public opinion. Under East European -14-
socialism, profit, however measured, was one factor, but it was only one. Other factors were
political and symbolic, such as "remaking nature". Frequently these political and symbolic factors
overrode all economic considerations. This is reflected in the statement of Soviet Academician
Fedorov made in the early seventies, and the position of Slovakia in its pleadings. Just as Fedorov
expressed the view that "Once a socialist society is established over the whole of the planet,
27
ecological crises will cease" , so the Slovak Memorial says that the Project presents a unique
opportunity to remedy the problems of a rapidly deteriorating and highly artificial river
28
landscape . Such pronouncements are based on the belief that socialist engineers designing and
constructing the Project could cure what the earlier (presumably capitalist) engineers, river trainers,
energy plant builders – or even nature itself – might have spoiled.
15. Whatever one thinks about the role of the State as an organizing power in remaking
nature and proving the superiority of planned economies, the Gab_íkovo-Nagymaros Project was
intended to be an investment, and a joint one at that. As market pressures began to be felt in the
region through the appearance of real prices and the collapse of the protective umbrella of Soviet
subsidies, changes became inevitable. They forced investors, including States, to consider the costs
of foregone alternatives and previously neglected costs, such as those relating to the environment.
16. The sponsors of the Original Project had to find ways of responding to emerging
challenges. The joint venture envisaged in the fifties and codified in the seventies involved not
only joint construction and ownership but also joint operation. The Treaty clearly required this,
among others by identifying the role of the government plenipotentiaries during the operation of
the Barrage System . The Parties were obliged by the 1977 Treaty and the regulations
implementing it to closely co-operate both in the day-to-day running of the Barrage system as well
as in responding to extraordinary or unforeseen circumstances.
17. And as I have already indicated, joint operation had to conform with environmental
requirements at the time of construction and during operation. A project meant to operate for
several decades cannot be free from the impact of subsequent regulatory norms. The Parties concur
27HR, Vol. 2, App. 3, at p. 131.
28
SM, para. 1.18.
291977 Treaty, Art. 3 (3) (b). -15-
on this. They agree that the operation of the Original Project must not have led to the impairment
30
of the Danube waters, including underground waters , and that compliance with the requirements
31
for the protection of nature ought to have been met throughout its lifetime . Disagreement may
remain over the extent of the obligations, but not as to the requirement that the joint investment had
to meet changing environmental expectations and obligations including those not envisaged at the
time of the Project's conception. In this way the 1977 Treaty provided for the continuous
integration of applicable environmental standards, including those imposed by subsequent treaties
and other rules of international law.
18. In connection with the object of the Treaty and character of the Project let me turn now
to another matter dividing the Parties which is of great importance. That is the question of whether
32
the sponsors of the Original Project intended to generate a territorial régime . Slovakia maintains
that the 1977 Treaty is a dispositive treaty33or a territorial treaty having the character of an
objective régime . Thereby it seeks to establish that Slovakia has some sort of in rem rights over
the water, that Hungary somehow alienated its sovereignty over the water. This is unsupported by
the Treaty. There is not a single provision in it which could justify this conclusion. Nor is the
conclusion supported by general international law, as Professor Crawford will show later this week.
19. To conclude my second theme, let me repeat that the character of the Project was a joint
investment, and one which had to be consistent with economic viability and environmental
protection. It was one which would be jointly operated. How far the Variant C reflects those
characteristics, we will see tomorrow.
IV. Relationship with other pertinent instruments
20. Mr. President, Members of the Court, I move to my third main point concerning the 1977
Treaty, namely its relationship to other agreements in effect between the Parties.
The related agreements
30
SR, para. 3.18.
31SR, paras. 3.32-3.34.
32SC-M, paras. 2.35-38, 2.48-52, HR, paras. 3.135-151, SR, para. 2.16-17.
33
SC-M, para. 2.38.
34SC-M, para. 3.39. -16-
21. Besides the 1977 Treaty as modified in 1983, the Project generated seven related
bilateral agreements between the Parties . Three of these concerned mutual assistance to be given
36
each other. They were the Mutual Assistance Agreement of 1977 as twice amended . The other
four agreements dealt with (1) the drafting of the Joint Contractual Plan, (2) customs, (3) the
regulation of the work of the plenipotentiaries and (4) border crossings in connection with the
construction and operation of the Barrage System. These were formal international treaties, signed
at governmental (at least ministerial) level. Their fate was linked to the Treaty and it is not in
dispute that if the 1977 Treaty terminated, so did they.
Other relevant agreements
22. A second circle of treaties comprise those relevant bilateral and multilateral treaties,
either pre-dating or post-dating the 1977 Treaty, which established obligations for the Parties to
this dispute independently of the Project. The most important is the 1976 Boundary Waters
Convention, others include the 1992 Biodiversity Convention. Their effect – together with the
obligations based on general rules and principles of international law – will be addressed later this
week.
The Joint Contractual Plan
23. Apart from these various agreements and other relevant rules and principles of
international law, there was a host of documents produced by the Parties with regard to the Project.
Many of these were typically socialist instruments combining public and private law elements with
political commitments. This was true, for example, of the Joint Investment Programme of 1973
which led to the elaboration of the Joint Contractual Plan a few years later. It was followed by
dozens of protocols and reports adopted in the Mixed Committee for Economic and Scientific-
Technical Co-operation or by the Government Plenipotentiaries or in the Joint Operational Group.
None of the Parties to this dispute thought it necessary to reproduce more than a few out of the
hundreds of the protocols and I need not refer to them any further. But the difference between the
Parties concerning the Joint Contractual Plan merits more detailed examination.
35They are listed in HM, para. 4.53.
36
HM, Vol. 3, Annexes 22, 29 and 30. -17-
37
24. Slovakia maintains that the JCP "appears to be a treaty in the full sense of the term" ,
that it may be "regarded as an element of the 1977 Treaty itself", and that a violation of the Plan
was equivalent to a violation of the 1977 Treaty . By contrast, in Hungary’s view, the JCP was
39
subordinate and ancillary to the 1977 Treaty , and was incapable of modifying its purposes and the
40
obligations it imposed . As a technical document – or series of documents, because it consisted of
a range of documents produced at different dates – the JCP reflected the specific role of a socialist
State as investor in a transborder economic venture.
25. To understand the status of the Joint Contractual Plan requires some familiarity with the
system of economic co-operation among socialist States. In planned economies no international
business transaction could be accomplished without "approval" of the Planning Office, other
competent State agencies and frequently the Communist Party itself. For its part the JCP was
41 42
elaborated by six private companies and endorsed by a low level bilateral technical committee .
Such a document is certainly not a treaty as envisaged by Article 11 of the Vienna Convention on
the Law of Treaties. That Article refers to national approval by the State and not to endorsement
by a technical supervising body. Slovak efforts to prove that the JCP was a formal treaty 43
contradict the facts. The JCP was not drafted by States but by companies, subject to private law. It
was not signed by national authorities having full powers to represent their State nor approved
according to the domestic rules applying to the approval of treaties. On the Czechoslovak side, the
normal signature of approval of any part of the JCP was affixed by Engineer Oblozinsky of the
44
Bratislava Water Construction Company – of whom the Court will hear more later. The content
37SC-M, para. 2.62.
38SM, para. 6.12.
39
HR, para. 1.16.
40
Ibid.
41Article 4 of the 1976 Agreement regarding the drafting of the Joint
Contractual Plan, HM, Vol. 3, Annex 18.
42Article 6, paragraph 3, of the 1976 Agreement regarding the drafting of
the Joint Contractual Plan, HM, Vol. 3, Annex 18.
43SC-M, para. 2.63.
44See SM, Vol. II, Annex 3, p. 34. -18-
of the JCP was not governed by public international law, but by private law governing business
transactions, valid at that time among socialist States.
26. Even identifying the “text” of the Joint Contractual Plan would be an insurmountable
task: the summary alone of the Joint Contractual Plan was divided into 31 volumes. The Plan itself
consisted of innumerable technical drawings, many thousands of pages which described designs in
constant re-evaluation and amendment according to the availability of construction material, labour
or the general state of the current five years plan. It was not a document but a filing cabinet, and a
filing cabinet of several companies, not of the State. In no sense was it a single determinate text. It
is no accident that at no point has Slovakia relied on any precise passage or paragraph of the “text”
of the Joint Contractual Plan in support of any disputed point, whether as to the agreed discharges
or its alleged provision for “underwater weirs”.
27. This is not to deny that the Joint Contractual Plan was a tool to implement the 1977
Treaty. Rather than a treaty governed by international law, concluded between two States acting
iure imperii and having continued capacity to modify the 1977 Treaty, it was a blueprint developed
by State-owned companies who implemented an investment decision of two socialist States, which
were the only possible investors under centrally planned economies.
V. The flexible character of the Project – at least until 1989
28. I turn to my fourth and final theme, the framework character of the Treaty and the
intended flexibility of its implementation.
29. Deviation from the time schedule and from the Joint Contractual Plan was rather the rule
than the exception, as indicated by the fact that 74 amendments to the Joint Contractual Plan had
45
been adopted before 1 January 1985 . The time schedule for putting the turbines into operation
was modified twice. One of them was reflected in the 1977 Treaty’s text, the second was not. This
shows that substantive changes could be adopted by the parties with or without formalizing it by
amending the framework treaty.
30. Slovakia agrees with Hungary that the 1977 Treaty “is a framework instrument, imposing
general obligations on the parties, with implementation being left to complementary and derivative
45
HM, para. 2.22. -19-
46
instruments” . It also concedes that the provisions of the 1977 Treaty could be – and had to be –
supplemented and adapted, in the light of experience, through the agreed provisions of the Joint
47
Contractual Plan . Nevertheless it asserts that the object of the Treaty, the construction of the
48
Gab_íkovo-Nagymaros Barrage system as a joint investment was “fixed ne varietur” . Hungary
agrees with this as long as it expresses the firm intention to proceed jointly and have an
economically viable, environmentally sound investment. But it can not accept the view according
to which modifications which were called for by the expected harmful effects of the Project were
prohibited, and could not have been introduced into the Treaty itself.
31. There is a contradiction here: Slovakia agrees with Hungary in the flexible nature of the
regulatory system and admits the evolutionary character of the Project up to 1989. That is dubbed
“implementation”, “supplementation” “adaptation” as long as the modification of the design or the
planned operational mode coincided with Czechoslovakia’s economic needs and interests.
However, when Hungarian economic and environmental concerns were presented as the grounds
for the suspension of work or the invitation for additional research, “adaptation” and
“implementation” was forgotten and Hungary was branded with the stigma of breach and violation.
32. Hungary believes that its suggestions in 1989 to review the Project from an
environmental point of view could have led to adjustments both in design and operational mode
compatible with the Treaty or acceptable by both Parties through the modification of the Treaty,
and would not have undermined the object and purpose of the Treaty as understood and
implemented thus far.
Conclusion VI.
33. Mr. President, Members of the Court, in conclusion let me summarize. -The 1977 Treaty
had two paramount purposes: the political, which called for the strengthening of fraternal
relations between Czechoslovakia and Hungary by enhancing their socialist integration, and the
economic, which aimed at economic benefits and viability consistent with applicable
environmental standards.
46SC-M, para. 2.58.
47
SR, para. 2.60.
48SR, para. 2.59. -20-
- The object of the Treaty was the construction and operation of the Barrage system under joint
control based on the agreement of the Parties. That agreement did not establish a territorial
regime but set the framework. With the passage of time and the increased importance of market
forces the construction and operation of the Project ought to have shown characteristics of a
reasonable investment: profitability and environmental sustainability.
- The Project was addressed by seven related international agreements, which shared the fate of
the 1977 Treaty. It was also subject to other relevant agreements, and to the rules and
principles of international law, and these for the most part continue to bind the parties to the
present litigation, whether or not the 1977 Treaty is in force.
- The Joint Contractual Plan was a tool of implementation, ancillary to the Treaty, linked to it
and to the other agreements. But it lacked the necessary precision in content, let alone the
minimum requirements of form, to qualify as a treaty on a par with the 1977 Treaty.
- This shows very clearly that the 1977 Treaty was a framework agreement to be implemented
primarily through the constant revision of the Joint Contractual Plan, but also, if necessary, by
the amendment of the Project timetable and, where necessary, of the 1977 Treaty itself.
However in Autumn 1989, in response to the expressed Hungarian concerns about the viability
of the Project, Czechoslovakia broke the tradition of flexible adaptation according to the needs
of the parties, and insisted on realizing the object ne varietur. That insistence denying
meaningful negotiations is at the root of the present dispute.
34. But before coming to these issues, something needs to be said about the concerns which
Hungary had in 1989, and continues to have, about the Original Project as a whole. I would ask
you, Mr. President, to call on Dr. Kern to introduce these issues. Thank you, Mr. President,
Members of the Court.
The PRESIDENT: Thank you Professor Nagy. I now call upon Dr. Klaus Kern.
Dr. KERN:
7. THE ORIGINAL PROJECT - SCIENTIFIC CONCERNS
Introduction -21-
1. Mr. President, Members of the Court, it is a great privilege for me to appear before you
today to present Hungary's submissions on science. As a river engineer specializing in river
restoration I am frequently asked the question: "To what extent can you rehabilitate river systems
which have been disturbed by human intervention?" For the Danube, the question is reversed:
how much disturbance can a river system endure without losing its essential ecological functions?
I am greatly honoured to be asked to contribute to a case before this Court with wide implications
for the affected Danube reach and other rivers as well.
2. Our presentation over the next hour addresses Hungary’s scientific concerns of the
Original Project. It will be divided into four parts, addressing the most important aspects of
Hungary's concerns. To consider the potential impact of the Original Project, it is necessary to
explain how it was to operate. I will address this in Part I of the presentation. In the second part
Professor Wheater will describe the threat to the system of bank-filtered wells which supply
Budapest with drinking water. In the third part he will explain the anticipated impacts in the
Szigetköz region. Finally, he will then briefly address, from a scientific perspective, the
deficiencies of the environmental impact assessment procedures which had been carried out before
1989. -22-
I. PLANNED PEAK POWER OPERATION
3. I begin by describing how the Original Project was to function in a mode referred to as
"peak power". This has particular environmental impacts requiring special attention. Before doing
so, it is important to make clear that the Hungarian position is not one of outright opposition to
hydro-power. It is not a question of being 'anti-dam' and 'pro-nature'. From an environmental point
of view, hydro-power at appropriate locations and in an appropriate manner is preferable to many
other sources of electricity generation. This is the case for parts of the Danube. In Austria, for
example, there was little protest against a new dam at Vienna-Freudenau, since environmental
impact assessment made it clear that it would not affect large floodplain areas or valuable side-
branch systems. It is not located in a protected area. By contrast, 10 years ago, Slovakia led
international opposition to a projected dam at Hainburg, just north of Bratislava, which was then
cancelled on environmental grounds . These were similar to those invoked by Hungary. Therefore,
each case must be addressed on its own merits.
4. No one disputes that the Original Project threatened damage to this part of the Danube
system. The issue in dispute is whether those impacts could be managed. We cannot in this oral
phase enumerate all impacts or potential hazards to the environment or landscape. The Court is
invited to refer to the written pleadings of Hungary for impacts on forestry, agriculture, fishery and
50
for deficiencies in earthquake design . We will focus on those impacts which are especially
significant: water resources and biodiversity.
5. In this regard, both the sensitive location of the Project and the proposed mode of
operation — peak power — threatened extraordinary damage.
6. To consider the implications of peak power it is necessary first to summarize how hydro-
electric projects work. If the water arriving at a power station is continuously discharged through
the turbines, it is operated as what is known as a "run-of-the-river" plant. In this case the maximum
capacity of the turbines will only slightly exceed average flow of the river, and larger floods will be
49HR, Vol. 2, App. 5, Sect. 5.
50
For example HC-M, Vol. 2. -23-
discharged over the weir.
7. Most hydropower stations in low-land rivers work on a continuous basis. Some, however,
store a portion of the daily discharge volume in a reservoir, releasing the water at higher flow rates
to generate electricity during times of peak energy demand. This is known as "peak-power"
operation. Peak-power production requires the construction of certain physical structures. It is
also subject to particular difficulties, and engenders certain risks and dangers.
8. To begin with, an operational reservoir is needed to store part of the daily flow.
Moreover, the maximum capacity of the turbines – that is to say the amount of electricity they can
generate – is determined by the planned peaking mode rather than by the river's natural flow rate.
The discharge capacity of the headrace and tailrace canals — their size — will correspond to the
maximum flow through the power station. Last, but not least, the artificial daily floods caused by
peaking require the construction of a second, downstream barrage. This barrage needs its own
operational reservoir to limit downstream disturbance to the river and its ecosystem.
9. No other Danube hydropower station in Austria or Germany works in "peak-power"
51
mode . The Original Project is unique. Where such systems do exist, for example in the Upper
Rhine or in the Rhône, they are designed to operate in a moderate peaking mode only and in
accordance with strict conditions established by international agreement between all States
3
concerned (Illus. No. 7.1). With 1,100 m /s, the average flow in the Upper Rhine is about half of
that of the Danube. A system of 10 barrages is operated together, as seen on the chart. It has been
agreed between France and Germany that the discharge released during peak operation must not be
more than 300 m /s above the natural flow . Because of the risks and dangers water levels within
the barrage system may not be reduced by more than 50 cm. Below the last barrage at Iffezheim,
the natural discharge must be re-established throughout peak operation. Similar restrictions are
established for peak operation in the lower Rhône . 53
51HC-M, para. 1.206.
52HC-M, para. 1.211.
53
N. Bordiec & A. Frézet (1986) "La gestion automatisée d’aménagement à
buts multiples - L’exemple du Rhône", La Houille Blanche, No. 6, 427-440. -24-
10. How does this compare with the Original Project? The system was designed to operate
54
in a large-scale peaking mode , unparalleled in Western Europe. The Dunakiliti-Hrusov Reservoir
was planned to have a volume of about 60 million m , able to store an entire day's flow.
11. The discharge capacity of the power canal and the turbines was to be two-and-a-half
55
times the mean annual flow of the Danube . This meant that nearly the highest flood in an average
year could be discharged through the turbines by-passing the floodplain forest, which would stay
dry.
12. The envisaged large-scale peak operation required a tailwater reservoir to compensate for
the daily flood waves that would be created. This was the function of the Nagymaros dam, as
56
acknowledged by Slovakia , turning a free-flowing river into a reservoir of about 120 km in
length, as seen on the chart, but with a drop of only 7 m.
13. The details of peak operation were not finally settled by Hungarian and Czechoslovak
57
experts. However, the magnitude of peak operation was fixed and dominated the design of the
entire system. For a run-of-the-river plant working on a continuous mode, four instead of eight
turbines would have been sufficient. Even for moderate peaking, the power canal could have been
half of its actual size. Thus, the envisaged magnitude of peak operation is evident from the
capacity of the structures and machinery. Different flow rates would have different peaking modes,
54Peak operation modes are based on investigations by the Technical
University of Budapest (1989) Hydraulic-Energy Characteristics of the
Gabcikovo-Nagymaros Hydrosystem for modified mode of operation ; authors:
V. Nagy and J. Ratky (in Hungarian); they were published in a conference
organised in 1993 by Slovakia: Karadi, G.M. and I.V. Nagy (1993) “Optimal
operation of the Gabcikovo-Nagymaros Hydropower System”, Proceedings of
the International Conference “The Gabcikovo-Nagymaros System - Intentions
and Reality”, Sept. 7-9, 1993, Bratislava. Characteristics of the peak
operation modes are given in HC-M, Vol. 2, Fig. 2.5 and HC-M, Vol. 4 (1),
Annex 6.
55HC-M, Vol. 2, Chap. 2.3.2.
56
SR, Vol. 2, p. 20 (2).
57
Karadi, G. M. and I. V. Nagy (1993) Optimal operation of the Gabcikovo-
Nagymaros Hydropower System, Proceedings of the International Conference
“The Gabcikovo-Nagymaros System - Intentions and Reality”, Sept. 7-9,
1993, Bratislava. -25-
with one or two daily peaks, as seen on the inserted box on the lower half of the chart. At low-flow
conditions the turbines at Gab_íkovo would be shut down for about 18 hours per day, and almost
all of the low-flow discharge would be stored in the upper reservoir.
3
14. A small residual flow of just 50 m /s would be left in the old riverbed and more than
5,000 m /s would be released at Gab_íkovo over 6 hours. This is more than five times the average
58
low-flow rate of the Danube .
15. How does peak operation affect the environment? As envisaged it would induce daily
water level fluctuations in the Nagymaros reservoir of up to 4-5 m (Illus. No. 7.2). It would affect
not just the tailrace canal as suggested by Slovakia . Up to 15 km of the abandoned Danube
60
channel and the lower side-branch systems, as well as the lower end of the Mosoni Danube would
be impacted. Contrasting with seasonal variations in surface and groundwater levels, these would
cause daily disturbance to aquatic habitats and destroy riparian areas along the entire length of the
Danube ridge. This type of damage can be seen in the draw-down areas of pump-storage reservoirs
where the banks are stripped of all vegetation. Riparian areas, however, represent transitional
habitats between land and water which are highly valuable, sheltering rare plant communities with
many endangered species , as Professor Vida explained. This fact has been deliberately ignored
62
by Slovakia .
16. Another, somewhat strange consequence of large-scale peaking is a daily change in the
flow direction of tributary streams and channels. A sudden rise in water levels due to peak
operation would force large quantities of water to flow several kilometres upstream in the old
63
riverbed of the Danube, as well as nearly 20 kilometres upstream the Mosoni Danube (Illus. No.
58
Ibid.
59
SR, Vol. 2, p. 64 (6).
60HC-M, Vol. 2, Chap. 2.3.2.
61
R. J. Naiman & H. Décamps (Eds.) (1990) The Ecology and Management of
Aquatic-Terrestrial Ecotones, UNESCO, Paris, and The Parthenon Publishing
Group, Carnforth.
62SR, Vol. 2, p. 59 (2).
63
HC-M, Vol. 2, Chap. 2.3.2. -26-
7.3). This "tidal effect" of peak operation will severely affect water quality in the Mosoni
Danube as Professor Wheater will explain.
17. How does peak operation affect aquatic habitats and biota? In short, very significantly.
Disturbance of aquatic habitats would result for large areas of the Nagymaros reservoir, including
the backwater reaches of the old riverbed and the lower end of the Mosoni Danube 65. Small water
insects, which are the main food source for fish, prefer familiar currents, and hide during the rising
flows which normally announce flood events. This behaviour is an important strategy for the
survival of aquatic fauna seeking to avoid drift. The daily fluctuations in flow velocity between
almost zero and nearly 2 m/s provide wholly inappropriate living conditions for the aquatic
67
fauna . It is therefore not understood how the tailrace canal, described by Slovakia as an artificial
structure, specially designed for peak operation, could offer new habitats for fish . It would be an
alien environment.
18. Since experiments on the scale of the Original Project have not previously been
attempted, comparable evidence from experience elsewhere is not available. However, it can be
said that peak operation in much smaller Alpine rivers has caused significant damage to the aquatic
fauna .9
19. What kind of impacts would be expected below the planned Nagymaros barrage? That
barrage would retain all coarse bedload. From the history of the riverbed around Szentendre
Island70, it is clear that further significant riverbed erosion would be very likely (Illus. No. 7.4).
64HC-M, Vol. 2, Chap. 3.3.2.2.
65
HCM, Vol. 2, Table 2.4.
66
Ibid.
67HCM. Vol. 2, Chap. 4.4.2.4.
68
SR, Vol. 2, pp. 64 (6), 85 (5).
69
O. Moog (1993) Quantification of daily peak hydropower effects on
aquatic fauna and management to minimize environmental impact. Regulated
Rivers: Research & Management , Vol. 8, 5-14. Also see HR, Vol. 3, Annex
4.
70Hydrological Yearbook Hungary. -27-
Riverbed deepening around Szentendre Island started in the 1940's after river training works were
carried out. Dredging of sediments started in the 1960's and was terminated in 1980, with the
exception of minor local dredging for navigation . Nevertheless, deepening of the riverbed
continued until recently, as seen on the chart by the drop of low-flow water levels at various river
gauges. After construction of the Nagymaros barrage, sediment transport would be restricted to
suspended load, and hardly any bedload would arrive in the
71HCM, Vol. 2, Chap. 2.2.2. -28-
branches around the Island. This would increase erosion and change the sedimentation pattern in
the channels .2
20. The power station at Nagymaros was designed to release a minimum flow of 1,000 m /s,
but would also produce a peak flow itself (Illus. No. 7.5). Under low-flow conditions it was
envisaged to increase the flow to more than 2,000 m /s for six hours a day . These peak flows
would be released into the free-flowing part of the Danube. This contrasts starkly with the
situation in the Upper Rhine where no such peak flows are permitted in the Franco-German treaty.
In addition to the lack of bedload, peak operation would further reduce the stability of the bed.
21. In conclusion, it is not possible to quantify exactly what the changes in riverbed level and
in the distribution of sediments would have been occasioned by the Original Project, but there
would have been significant changes in the tailwater reach between Nagymaros and Budapest.
There is a sufficient basis for concluding that sediment disturbance by the construction and
operation of the Nagymaros Barrage would create real risks, risks which amply justify Hungary's
desire for further investigation. In particular, they would endanger the drinking water well groups
on and around Szentendre Island which supply the city of Budapest. This phenomenon will now be
treated in more detail by Professor Wheater. Mr. President, Members of the Court, I would like to
thank you for your attention. Mr. President I ask you to call on Professor Wheater to continue, I
thank you.
The PRESIDENT: Thank you, Dr. Kern. Professor Wheater, please.
ProfeWssorEATER:
II. BANK-FILTERED WELLS AND THE BUDAPEST WATER SUPPLY
22. Mr.President, Members of the Court, it is a great honour to address you for the first time.
72
HCM, Vol. 2, Chap. 2.3.2 and 3.6.5.2.
73HC-M, Vol. 2, Chap. 2.3.2. -29-
23. My presentations will address the following issues. First, I will describe the role and
functioning of bank filtered wells and the threat posed to them by the Original Project. Second, I
will describe the anticipated impacts of the Original Project on the Szigetköz region, in particular
on levels and quality of groundwater, on quality of surface water, and on the region's biodiversity.
Third, I will comment on the lack of any adequate impact assessment prior to 1989.
a) Bank-filtered wells
24. One of the basic obligations of any State is to maintain the drinking water supply available
to its citizens. In the 1980s it became apparent that the quantity and quality of the supply to
Budapest – most of which comes from bank-filtered wells – was vulnerable to the effects of the
Nagymaros dam, which would also affect water supplies to towns upstream.
25. Approximately 85% of the Budapest water supply capacity comes from 758 bank-filtered
wells. The supply capacity is over one million cubic metres per day. Three quarters of the bank-
74
filtered supplies are to the north of the city, mainly in the major well fields of Szentendre island ,
which you will visit next month. These are just downstream of the Nagymaros dam site, as shown
here (Illus. No. 7.5). Upstream of the Nagymaros dam site, bank-filtered wells are also important,
75
for water supply to local towns and villages .
26. Bank-filtered wells are a long-established method of abstracting river water for drinking
water supply (Illus. No. 7.6). They are sited next to rivers and abstract water from deposits of
underground alluvial material connected to the river. The wells draw on the river water, and the
passage of the water through the alluvial sands and gravels provides a natural filtration which is
highly effective in removing pollutants. This diagram illustrates the operation of a typical bank-
filtered well. The quantity produced by a well depends on 4 factors: river water level, the thickness
of the filter layer, its connection to the river, and the physical properties of the filter layer.
Additionally, the removal of contaminants depends on the length of the filter pathway and the
properties of the filter layer. Alter these characteristics, and you risk altering the yield and quality
74HC-M, Vol. 2, Sect. 3.6.1, p. 105.
75
Ibid. -30-
of the well.
b) The threat to bank-filtered well systems
27. The primary concern for bank-filtered water wells comes from changes in river bed
sediments. River-bed erosion can reduce river water levels and hence the available well yield.
More seriously, removal of the gravel layer will cause loss of water quality protection and can
threaten the well supply in its entirety.
28. Where deposition of fine sediments (e.g. silt particles) occurs, important chemical changes
may result from the degradation of organic material. The degradation consumes oxygen, which can
result in the dissolution of iron and manganese, and the production of ammonium. These may
render the water unsuitable for drinking, and can cause bacterial slimes which clog the well
screens.
29. These are the basic principles. Let me now turn to actual Hungarian experience concerning
degradation of well water quality.
30. The Surány Waterworks, Szentendre island, immediately downstream of the proposed
Nagymaros dam, illustrates the degradation of water quality that can arise due to sediment
76
deposition (Illus. No. 7.7) . Water quality problems began in the 1970’s, and a detailed study was
carried out in the mid-1980’s to identify the cause. This showed that two troughs in the river bed
had filled with fine sediment, the degradation of which led to severe water quality deterioration,
commencing in 1974-1975. By 1984 ammonium levels in one well reached 90 times EC guide
levels for drinking water, while manganese levels were 200 times EC guide levels. The problems
persist today .
76HC-M, Vol. 2, Sect. 3.6.3.
77
HC-M, Vol. 2, fig. 3.22 and figs. 3.25, 3.27. -31-
31. Similar problems exist around Nagymaros. Deterioration of water quality began in the
early 1980s, and several wells had to be removed from service as drinking water standards were
78
exceeded . Again, the adverse changes were due to sediment deposition and degradation.
32. The Hungarian experience is that bank-filtered wells are extremely sensitive to changes in
bed sediments, that the processes of chemical degradation can readily lead to chemical
concentrations which greatly exceed drinking water limits, and that effects can last for decades.
33. Sediment degradation is not the only problem. Erosion of the river bed is a further concern.
This lowers river water levels and removes the protective filter layer. The lowering of river levels
reduces well yield – that is to say the amount of water available. More important is the effect of
gravel removal on the filter performance, and the possibility of complete removal of the bed
filtration layer.
78
HC-M, Vol. 2, Sect. 3.6.3.2. -32-
34. The history of bed degradation at Szentendre has already been discussed by Dr. Kern. The
long-term decline in bed level has now been halted by strict conservation measures. The current
state of river bed gravels shows complete removal of the gravel bed in several areas. The Budapest
Waterworks current estimate of the decrease in well capacity due to dredging is approximately 30%
79
of the total water resource potential of this Nagymaros to Budapest reach. Thus a major impact on
Budapest’s water supply has already occurred, as a result of river bed degradation. The
vulnerability is clear.
c) Anticipated Impacts of the Original Project
35. Having considered the evidence of water quality degradation and impacts of bed changes, I
turn now to the expected impacts of the Nagymarosdam. I begin with wells in the back-water reach,
upstream of the proposed Nagymaros dam; I thenconsider the wells which supply Budapest.
79 HC-M, Vol. 2, Sect. 3.6.4, p. 115. -33-
1) Impacts of the Nagymaros Dam on bank-filtered wells in the backwater reach (Gönyû to
Nagymaros)
36. As with all dams, deposition of fine sedimentis anticipated in the backwater reach behind the
dam. This is due to reduced flow velocities. It has important consequences for bank-filtered water
supply. The long-term effect of sediment cloggingis estimated to reduce well yields by up to 40% . 80
Calculations of the impact of sedimentation on well water quality show increased levels of iron and
manganese . On this the parties do not disagree, since Slovakia expected clogging, sediment
82
degradation and groundwater quality de terioration in the Cunovo reservoir . The Court will hear later
of the observed impacts in that reservoir, but it islevant to note here that predicted chemical changes
83
are actually being observed today . The experience with Variant C c onfirms the main fears associated
with the Original Project.
37. There can be no doubt that Nagymaros posed a potentially serious threat to these upstream
water supplies. Clogging would lead to loss of yield, and changes in water quality would render the
supply unsuitable for drinking.
2) Impacts of the Nagymaros dam on Budapest water supply
38. I turn now to the downstream effects of Nagm y aros. The water supply to Budapest is obviously
of paramount national importance. Two thirds of Budapest’s total drinking water comes from the bank-
84
filtered well fields between Nagymaros and Budapest . The sensitivity of this supply has been
demonstrated. Any threat to this supply is a threat to a vital national resource.
39. The precise impact of the Nagymaros dam on downstream river bed sediment patterns is
difficult to predict. Different calculation methodscan produce order-of-magnitude differences in river
80
HC-M, Vol. 2, Sect. 3.6.5.1, p. 116.
81
Ibid.
82
SR, Vol. II, Part II, p. 166.
83
SR, Vol. II, Part II, p. 142.
84
HC-M, Vol. 2, Sect. 3.6.1, p. 105. -34-
bed changes . Sticking to first principles, however, dams invariably retain most of the coarse
sediments normally associated with bed-load transport. The loss of these sediments to the channel
downstream induces downstream scour, an effect already observed downstream of the Variant C
dam . The magnitude of this downstream bed erosion is uncertain. Three things are certain: (1) bed
changes would definitely occur (2) these changes would have been aggravated by peak power
operation, and (3) these changes would threaten furthreremoval of the fragile filter layer on which the
bank-filtered wells depend. Additionally, irregular pulses of fine sediments would be released by the
88
dam and could settlein pockets downstream . This would cause severe local degradation of well water
quality, as already observed .
40. We have noted that, in its discussion of ban-filtered wells downstram of Nagymaros, Slovakia
agrees that changes in the f ilter layer pose potential threats The issue for the Court, then is not
whether a threat existed, but whether the threat jutsified further investigation, and then action. Slovakia
makes no reference to the expected river degradation caused by Nagymaros. With a strange logic, it
cites as evidence of lack of Hungarian concern a repotrindicating that “special attention” must be paid
91
to river bed problems . However, it conveniently overlooks the previous paragraph of that report,
which states that “The riverbed is expected tdegrade with the operation of the barrage system.”2
85
HC-M, Vol. 2, Sect. 3.6.5.2.
86
Rákóczi, L. and Sass, J. (1995) Changes of the Channel of the Hungarian
Upper Danube and of the Side River Ar ms of the Szigetköz upon putting the
Dunacsúny I. River Barrage into Operation. Vízügyi Közleménye,kVol. 77, pp.
46-70 (in Hungarian).
87 HC-M, Vol. 2, Sect. 3.6.5.2.
88 HR, Vol. 2, Sect. 4.4.2.
89 HC-M, Vol. 2, Sects. 3.6.3.1, 3.6.3.2.
90 SR, Vol. 1, pp. 280-282), SR, Vol .1, para. 12.03, p. 280.
91 The SR (Vol 1, p 282) quotes a Somlyody et al. report (HC-M, Vol. 4, p. 576)
92 Ibid . -35-
41. The effects which I have described clearly posed a major threat to the water supply of
Budapest. In response to concreteHungarian evidence of yield redutcion and water quality degradation,
Slovakia states simply that “there is no support for such a conclusion” . It consistently ignores the
Hungarian science, much of which I have brought to your attention this morning.
III. NTICIPATED IMPACTS IN THE S ZIGETKÖZ R EGION
42. In this second part of my presentation, I will now address the anticipated impacts of the
Original Project for the Szigetköz Region, the importance of which was explained yesterday by
Professor Vida.
43. The assessment of impacts on the region raises complex issues, in particular for the chemical
and biological responses which would have resulte d from physical changes, and the consequential
implications, for example, for surface water and groundwater quality, and for ecological response.
Knowledge and awareness of these matters has increased over the last decade. With the benefit of
hindsight it is clear that these complexities were either ignored or dismissed on superficial analysis in
the impact assessments of the Original Project. Theri importance has now been recognised, at least by
Hungary and international scientists.
44. Nevertheless, uncertainties remain, and we should not be frightened to say so. The best
scientific predictions of risks will always containelement of uncertainty. Slovakia’s view that “The
94
kind of ‘uncertainty’ invoked by Hungary does not exist” is simply not tenable. Slovakia pours scorn
on Hungary’s recognition that complex, long-termenvironmental impacts, which may represent major
risks to resources of national and international significance, come with a degree of uncertainty . It
adheres to the curious view that monitoring over three years is sufficient to detect long-term change,
and that monitoring can guarantee that all adverse effects can be managed 96. I hope the Court will
93
SR, Vol. II, p. 27.
94
SR, Vol. I, para. 1.15, p. 6.
95
SR, Vol. I, para. 1.16, p. 6.
96
SR, Vol. I, paras. 1.16-1.18, pp. 6, 7. -36-
recognise that to accept uncertainty is not to deny risk, or to limit the scope for reasonable, prudent
protective measures.
45. In this context, and given the limited time ava ilable, I cannot discuss in detail all aspects of the
97
impacts of the Original Project in the Szigetköz. These are given in Hungary’s written pleadings . I
will review the flows fixed in the Original Pr oject plans and then focus on three issues: (1)
groundwater levels and quality, (2) surface water qualityand, (3) ecological response. I will focus on
perceived risks, since the Original Project was notcompleted. Tomorrow, I will consider the observed
data on the impacts of Variant C.
46. The Joint Contractual Plan fixed dischargesto the Danube river channel, the Mosoni Danube
and to the floodplain (Illus No 7.8). To maximise electricity generation it was agreed that the residual
3 3
discharge into the Danube be fixed at 50 m /s. In winter this would fall to 20 m /s or less. During the
vegetation season it could increase to up to 200 m/s, still only 10 % of the natural average flow. And
3 98 3
far less than the 800 m /s which the EC recommended , or the 600 m /s which Slovak government
99
advisers considered necessary. Lesser amounts would go to the Mosoni Danube and the Hungarian
side-branch system.
47. The operational plans for the water distributo i n only allowed flood flows in the main riverbed
once the discharge capacity of the turbines was exceeded. And only above the even higher level of
3
6,500 m /s would the side-branches in the upper and middle Szigetköz receive additional water.
100
Inundation of large areas of the active floodplain owuld occur on average less than once in 20 years .
48. The wetland ecosystem would thus be deprivedof its most essential elements – water, and the
flow of water at varying levels. This would ha ve significant impacts on groundwater levels and on
groundwater quality.
97
HC-M, Vol. 2.
98
HM, Vol. 1, Ann. 15, Sect. 3.2.
99 HM, Vol. 1, Ann. 15, Sect. 3.1.2.
100HC-M, Vol. 2, Chap. 2.3.2. -37-
Groundwater a)
49. To consider the impacts on groundwater, one mustfirst describe the situation prior to construction
activities. The video identified the large alluvial aquifer in the Szigetköz and Zitny Ostrov created by
the Danube (Illus. No. 7.9). The volume of water in this aquifer, on the Hungarian territory alone, is
estimated to be 5.4 cubic kilometres 101. The aquifer was recharged mainly by the Danube channel;
recharge from rainfall is very limited.02 Natural processes of recharge maintained good groundwater
103
quality , but the system is very vulnerable to change.
50. Groundwater flow velocities – and hence travel times – are extremely slow. It will take many years
for pollution to move through the aquifer. A useful result of atmospheric bomb tests, carried out in the
late 1950’s and early 1960’s, is that tritium was deosited and now can be used as a time marker. The
extent to which tritium has penetrated the Szigetközaquifer can be used to trace the contour of 30 year
104
travel time from the Danube (Illus. No. 7.10). This shows that rates of travel are less than 300 metres
per year: in other words, a pollutant may take 10 years or more to move just 3 kilometres. And such
pollution will at best be long-lived; at worse, irreversible.
51. The average groundwater levels mask a highly dynamic groundwater response based on Danube
floods (Illus. No. 7.11). As shown in the video, when the Danube levels increase, the river water
recharges the aquifer. When leve ls decrease, there is a local drainage return flow to the Danube
channel, providing a natural flushing of the gravel bed.
106
52. As can be seen from this diagram, historical decreases in groundwater levels (so frequently
101
HC-M, Vol. 2, Sect. 3.4.1.1.
10Ibid.
10HC-M, Vol. 2, Sect. 3.5.1.
104
HC-M, Vol. 2, Sect. 3.5.1 and Fig. 3.7.
10Ibid.
10HC-M, Vol. 2, Sect. 3.4.1.1. -38-
mentioned in the Slovak pleadings ) have been very limited in the Szigetkoz.
53. In fact, the major decreases in groundwater levels near Bratislava are clearly shown by
Slovakia (Illus. No. 7.12) to be a result of local groundwater extraction.
54. The natural processes of groundwater recharge to the Szigetkoz from the Danube used to result
in excellent quality groundwater. The infiltrated water was sufficiently rich in oxygen to oxidise the
low amounts of degradable organic material. As aresult chemical constituents such as iron, manganese
and ammonium have only been present in low concentrations. 109 This high quality of recharge water
ensured that most of the Szigetkoz had good gr oundwater quality. Pollution has been limited; poor
110
quality water is generally localised anderstricted to the upper 20m of the aquifer .
55. The Szigetkoz aquifer thus represents an immense resource of high quality water of national
strategic importance. The 1978 Joint Contractual Paln posed a clear threat to that resource.
1) Groundwater levels
56. By removing 95% of the water from the Danube the 4 metre reduction in river levels would
111
have caused a general reduction in groundwater levels and change in groundwater flows . The
Dunakiliti reservoir would have caused locally incr eased groundwater levels, and, at least initially,
become the main source of groundwater recharge. However, the operation of the reservoir would be
107
E.g., SM, Vol. 1, para. 1.58; SC-M, Vol. 1, para. 7.81 and Illustration
No. CM-5; SR, Vol. 1, para. 13.05; SR, Vol. 3, Chap. 1, Part 4.
108
SR, Vol. 1, Illustration R-11.
109HC-M, Vol. 2, Sect. 3.5.1.
110
Ibid.
111HC-M, Vol. 2, Sect. 3.4.2.1. -39-
accompanied by deposition of fine sediments, whic h would in the long term reduce the recharge
through the bed .12
57. An attempt has been made to quantify these effects using groundwater simulation modelling
113 3
(Illus. No. 7.13) . The estimated impact on average groundwater levels for a flow of 50m/s shows a
rise of 3 m close to the dam. However, over most of the area, groundwater levels fall, in some cases
2 2
very significantly. An area of 20 kmsuffers a decline in levels of more than 3m; 75 km has a decline
2 114
of more than 2m. The total area affected by reduced levels is over 300 km . An increased Danube
flow of 200m /s would bring only minor improvements: the area affected would be just under 300
km . More pessimistic (but plausible) estimates of clogging show further reductions in groundwater
115
levels .
58. The effect on average conditions masks other important impacts. The variability of
groundwater levels would largely be lost with implementation of the Original Project. The
consequences would be a greater reduction of peakgroundwater levels than the average levels which
have been illustrated. This natural variability iessential for the functioning of the natural system, e.g.,
floodplain vegetation, maintenance of self-cleansinggravel river bed conditions, and oxygenation of
the groundwater.
59. The estimated effect of the reduced groundwater levels on the natural processes of sub-
irrigation is shown here (Illus. No. 7.14). Natu ral sub-irrigation depends on the groundwater table
reaching the fine soils which overly the alluvial aqufir. It provides an essential soil moisture supply in
dry periods. Dam construction would reduce the area benefitting from this process by approximately
116
one half .
60. Apart from the loss of natural sub-irrigation, long-term changes to the soil profile, including
112
Ibid.
113HC-M, Vol. 2, Sect. 3.4.2.1; Vol. 5, Plate 3.11.
114
HC-M, Vol. 2, Table 3.4.
115
HC-M, Vol. 5, Plate 3.16.
116HC-M, Vol. 2, Sect. 3.4.2.1; Vol. 5, Plate 3.12. -40-
development of carbonate accumulation layers, may o ccur. Again, Slovak soil scientists share this
concern . It is just one of many aspects of the Original Project not considered prior to 1989.
2) Groundwater quality
61. The Original Project raises equally serious concerns for groundwater quality. These have been
118
consistently dismissed by Slovakia in its written pleadings , despite the evidence described by
Professor Carbiener and acknowledged by Slovak and international scientists .
62. Under the Original Project the main sources of groundwater recharge would become the
Dunakiliti reservoir and the network of side-arm channels. The reservoir would inevitably cause
deposition of fine sediments over its bed. As I explained earlier, these fine sediments contain organic
material which will degrade, consuming oxygen. If chemically-reducing conditions occur, iron,
manganese and ammonium will be released into solu tion, and the quality of groundwater recharge will
120
deteriorate . These processes have been clearly demonstratedfor bank-filtered wells. In fact, there is
considerable international experience of such problems 12, as reviewed by Professor Carbiener,
including dams further upstream in the Danube.
63. For example, intensivereservoir at Altenworth, in Austria, wree undertaken in the 1980’s (Illus. No. 7.15.)The observed changes due
to the damincluded exactly the same aspects of concern he re for the Original Project: deposition of
organic-rich sediments in the reservoir, infiltration from the reservoir to groundwater through these
organic-rich sediments, a decrease in groundwater fluctuations (and hence oxygen supply), and the
reduction of inundations of oxygen-rich surface water. These changes caused severe degradation of
117
HR Vol. 2 pp. 60-61.
118
E.g., SR, Vol. II, p. 43.
119E.g. SR, Vol. II, Part II, p. 166.
120
HC-M, Vol. 2, Sect. 3.5.1.
121
HC-M, Vol. 2, Sect. 3.5.2.1.
122 Documented in detail by Professor Nachtnabel and colleague, from the
Agricultural University of Vienna (Hary and Nachtnabel, 1989). -41-
groundwater quality .23
64. A key point to note is that these processes were not instantaneous, but emerged only after
several years. This illustrates an obvious fallacy of Slovak arguments that monitoring is somehow
adequate to prevent environmental da mage. This is clearly incorrect for the Altenworth experience. It
is incorrect for processes of long-term change in general.And it is incorrect in respect of the Original
Project.
124
65. Hungarian analysis indicates that degradation of groundwater quality would result from
125
recharge from the Dunakiliti reservoir. Thisiew is confirmed by Slovak scientists .
66. The other main source of groundwater recharge would be the side-arm channel system in
the Szigetkoz. A similar set of concerns apply here . As will be described later, the side-arm channel
system contains fine organic sediments. Moreover, the underlying geological sequence consists of a
127
complex interleaving of alluvial gravels and fine sediment materials . It follows that infiltration
through this sequence is subject to the same concerns for chemical degradation. The poor quality of
128
infiltrating water has been clearly identified from monitoring programmes . I will describe these in
more detail when I consider the impacts of Variant C.
67. To summarize, the long-term implications for groundwater quality of the Original Project
are as follows. Water of degraded quality from the reservoir bed and the side-arm system 129would
123 Hary and Nachtnabel write that “In the northern floodplain, for
which extensive data is available, the groundwater quality indicates
an oxygen-depleted or oxygen-free zone. Simultaneously, increased
iron and manganese concentrations were found after a period of delay
of a few years after the power station construction.”
124 HC-M, Vol. 2, Sect. 3.5.2.2.
125 SR, Part II, p. 166.
126
HC-M, Vol. 2, Sect. 3.5.1.
127 HR, Vol. 2, Sect. 7.3.2.
128 HC-M, Vol. 2, Sect. 3.5.2.3.
129
HC-M, Vol. 5, Plate 3.15. -42-
replace high quality recharge water from the Danube channel, and progressively replace the high
quality groundwater of the Szigetkoz.
68. Such changes render the water unsuitable fo r drinking. They would seriously degrade a
major and vital water resource. In addition, th e loss of oxygen, together with the reduction in
groundwater levels and loss of their natural variab ility, would have very serious implications for
agriculture and the natural ecosystems.
b) Surface water quality
69. Let me turn now to the impacts on surface water quality.
70. The Danube, as is the case for most major European rivers, is used as a sink for industrial,
agricultural and domestic wastes. The Danube water qulaity in Hungary thus contains a wide range of
pollutants. It is also vulnerable toaccidental pollution generated upstream.
71. The bacteriological water quality refl ects the organisms commonly associated with
untreated sewage 13. Hence the entire Hungarian reach fro m Budapest upstream is unsuitable for
bathing. Heavy metals in the water, for exampl e, mercury, lead, cadmium, can exceed applicable
limits . The river sediments contain heavy metals, and organic pollutants. These contaminants are
mainly associated with the fine sediments, for which average concentrations of heavy metals exceed
132
limit values .
72. Long-term changes have taken place. Porgressive deterioration occurred from the 1960’s to
133
the mid 1980’s . Dissolved oxygen is an essential requirement for most aquatic organisms, and a
common measure of river pollution is the rate of consumption of oxygen by biological and chemical
oxidation processes. This period saw a near doubling of this Biochemical Oxygen Demand (BOD) and
130 HC-M, Vol. 2, Sect. 3.3.1.3.
131 HC-M, Vol. 2, Sect. 3.3.1.4.
132
HC-M, Vol. 2, Sect. 3.3.1.5.
133
HC-M, Vol. 2, Sect. 3.3.1.1. -43-
a dramatic increase in the nutrients nitrogen and phos phorus (by a factor of ten in the latter case).
However, in the last decade, progress has been made in reducing emissions, and hence there has been
improvement in some water quality indicators .134
73. One effect of the construction of dams fu rther upstream has been that concentrations of
suspended solids decreased by about half between the late 1950s and the late 1970s. This gives
improved water clarity and light pene tration. Together with the increased nutrient levels, this provides
the classical conditions for enhanced growth ofalgae and higher aquatic plants, commonly known as
eutrophication. Algal growth was no longer limited by nutrient availability, and a ten-fold increase
135
occurred in the indicators of algal population, such as algal counts, algal biomass, and chlorophyll-a .
74. Eutrophication can have many adverse effect s. These include fluctuations in dissolved
oxygen levels that can lead to fish kills, enhanced sedimentation due to decaying organic matter and
changes to many aspects of river ecology. A changein the composition of algal populations can result
136
– for example, leading to the persence of toxic blue-green species . Under particularcombinations of
meteorological and hydrological conditions (that is, warm weather, high incident radiation and low
flows) massive growth of planktonic algae can occur, known as algal blooms, leading to severe
problems of oxygen depletion.
75. This is the background. I will now turn to the anticipated impacts of the Original Project,
noting that water quality was almost entirely overlookeidn the planning for the Original Scheme. I will
give just two examples of the impacts.
76. A simulation of the effects of the upstream reservoir in the Original Project on algal growth
137
at Szap is shown in this diagram in blue (Illus. No. 7.16) . Inter-year variability is marked, but the
average result is a 60 per cent increase in biomass. Such increases in biomass create a load of organic
134 By 4-7 per cent/year in the Rajka-Budapest stretch, 1986-1992.
135 HC-M, Vol. 2, Sect. 3.3.1.2.
136
Ibid.
137
HC-M, Vol. 2, Sect. 3.3.2.3, Fig. 3.6. -44-
material which consumes oxygen. This effect can q eual or exceed that of the added sewage effluent in
the Rajka-Budapest reach. Hence, with the Original Project eutrophication is expected to be a major
problem, particularly in the reservoirs and backwa ter reaches, and one which cannot be solved by
wastewater treatment alone.
77. A second example of water quality impacts concerns the effect of peak operation on water
138
quality in the Mosoni Danube . This river receives wastewaterfrom Gyor which reduces dissolved
oxygen. The effect of peak power operation would be to introduce the daily flow reversal described
earlier. This would significantly increase residence timefsor the water. This effect was not considered
139
at all until 1989 , when it was shown that low dissolved oxygen values could occur, and complete
oxygen loss might even be expected with catastrophic results for the fish population and the river
ecology in general.
78. Finally in this part, I turn to the ecological impacts.
79. How would the flow régime impact on aquatichabitats? The planned residual discharge of
3
50 m /s in the Danube riverbed would have resulted in a drop of surface water levels of about 4m,
reducing by half the extent of the wetted perimeter ofthe river. The remaining small river would differ
considerably from the real Danube in its ecologicalfunctions. Flow velocities would drop to one-third
their original rate . The lower half of the abandoned channel would be a backwater reach impacted
by peak operation as explained before.
80. This imposed régime would result in a complete change of sediment patterns in the lower
half of the main channel of the Danube. Large areas of the riverbed would experience deposition of
sand and silt. Clean coarse gravel, which used to dom inate sediments in the Danube channel, would be
limited to small areas only.
81. A smaller water volume, reduced flow velo cities and changed sediment pattern would be
138 HC-M, Vol. 2, Sect. 3.3.2.2.
139
Somlyody et al. (1989).
140
HC-M, Vol. 2, Chap. 2.3.2. -45-
141
detrimental to the aquatic fauna . Those fish which are typical for flowing rivers and need spawning
areas with clean gravel, or those which depend on side-branches connected to the free flowing river,
would be restricted to small areas and be reduced in diversity and abundance. Fish populations are
indicative of what would happen to the aquatic flora and fauna in general: a shift in species
composition would occur in favour of stagnant-w ater-preferring species. This would represent a
distinct degradation of the natural value of th is Danube reach. These impacts were predicted by
142 143
Slovak fish experts and accepted by Slovakia for the Original Project .
82. In addition, the Danube channel would suffer from its isolation between the two reservoirs
and by being cut off from its floodplain habitats. Whee floodplain habitats are connected to the river,
organic matter is washed out in the running water during inundation, providing food for the benthic
invertebrates and the river fish. These connected haitats are also used by numerous fish species from
the main channel as spawning and nursery areas aswell as for refuge during high spates or accidental
pollution events in the main riverbed. I would like torefer back to Professor Carbiener who explained
the ecological processes connecting the river andits floodplain which are essential for preserving and
sustaining the very character and substance of the riverine ecosystem.
83. The branch system itself contained a broad diversity of aquatic habitats. The construction of a
supply canal (as a so-called remedial measure) woudl require connecting all side-arms and closing the
entrances to the main channel. The resultant cons tant flow would level out the differences in the
physical-chemical properties (that is temperature, nutrients, oxygen content). The aquatic flora and
fauna would lose its diversity and a few specieswould be favoured and reachhigh abundance. Similar
144
processes were observed in the Upper Rhine , and were reported yesterday by Professor Carbiener.
141
HC-M, Vol. 2, Chap. 4.4.2.
142
Holcik, J., I. Bastl, M. Ertl and M. Vranovsky (1981), Hydrobiology and
icthyology of the Czechoslovak Danube in relation to predicted changes after
the construction of the Gabcikovo- Nagymaros River Barrage System. P ráce
Lab. Rybár Hydrobiol. 3: 19-158.
143 SR, Vol. 2, p. 61(6), 85(3), 85(4).
144
Krause, W. & G. Hügin (1987) Ecological Effects of the Management System
of Connected Side Branches (demonstrated by the example of the regulation of
the side branches of the river Rhine). Natur und Landschaft 62(1):9. HC-M, -46-
More water fed into the branches would not cure the problem as suggested by Slovakia 14. It would
make it worse. A more detailed discussion of reme dial measures will be presented tomorrow in the
context of the impacts of Variant C.
84. How would the Original Project impact on terrestrial habitats? The impacts on the wetland
ecology would be quite similar to the ones observed with the operation of Variant C, which will be
described later. Groundwater levels would drop, perventing capillary rise of groundwater in large areas
and resulting in a partial desiccation of valuab le wetland habitats. Species of drought-tolerant
vegetation communities would eventually replace hydrophilic species – a process which may take
many years or decades but which is currently observd ein connection with Variant C by the invasion of
weeds. In those areas where the side-branch watersupply would raise groundwater levels permanently
to a high level, the species composition of the previous wetland plant communities – dependent on
146
water table fluctuations andniundations – would change to tho se characteristic of marshy land .
IV.L ACK OF INTEGRATED ASSESSMENT
85. My colleague, Dr. Kern and I have outlined justsome of the concerns for the Original Project.
There is no doubt that the Project had the potential tocause very significant environmental impacts. In
this third, and final part of my presentation, I would like to comment briefly on scientific aspects of the
impact assessment for the Original Project. Later thismorning Ms Gorove will explain why the efforts
at environmental impact assessment relating to the Oriignal Project were deficient from the perspective
of international law.
86. Clearly, a substantial effort was made by bothsides to investigate some aspects of the perceived
environmental risks of the Original Project. Unforut nately in these matters effort alone is not enough.
When Hungary suspended work on its part of the Oir ginal Project, in 1989, the studies were inadequate
for proper decision-making: they were narrowly define d, incomplete, and insufficient in a number of
key areas and in particular, there had been no attempt at an integrated technical assessment, let alone a
comprehensive Environmental Impact AssessmentT . here was therefore no scientific basis for assessing
Vol.. 4(2), Ann. 15.
145
SR, Vol. 2, p 72(5).
146 HC-M, Vol. 2, chaps 4.4; HR, Vol. 2, chap 5.1. -47-
the extent of the impacts.
87. Permit me to consider – briefly – the scope and content of the Slovakian studies which, it is
claimed, demonstrate the extensive attention givn to environmental assessment. Such studies as have
147
been produced to the Court merely rein force Hungary’s concerns. For example, the concerns of
148 149
Slovak soil scientists are clearly demonstrated , and the incomplete nature of associated research . It
is said, among other things, that “The chemical properties of farmland soils and their variability and
dynamics have to be assessed”, “Remedial measur es must be proposed”, “It is necessary... to
implement long-term soil research methods”. All this from a 1993 Report.
88. Ms Gorove will discuss the 1985 Hungarian “Environmental Impact Assessment”. 150From a
scientific perspective, it contained numerous omissions and was subject to major inadequacies. For
example, the key processes of sediment degradation and their impact on quality of groundwater were
ignored, despite the international experience of which you have heard. River water quality modelling
neglected the impacts of nitrogen and phosphorus cycels and the effects of peak power operation. The
dramatic potential impacts on the water quality of the Mosoni Danube had not been evaluated. And
external review led to concern that scientific approach lacked objectivity.
89. The need for, and lack of, an integrated assessment was clearly identified by EC and Slovak
147 HR, pp 60-62.
148
“The construction of the Danube Ba rrage System constitutes a significant
intervention into the natural environmentof the region.....In a large part of this
area, changes in the groundwater level en tail the modification of the regimes of
farmland, modifying the characteristics ofagricultural soils, and changes in the
levels of high mineral content groundw ater may accelerate the accumulation of
salts in the soil or farmland profiles ...... Given the manifold nature of the
particle composition of farmlands and the soil, and the differences in the depth
and salt content of groundwaters, we haveto expect a wide range of changes in
the properties and transport characteris tics of farmland soil.” (Rehak et al.,
1993). HR, Vol. 3, Appendix 7, Part 2.
149 HR Vol. 3, Ann. 7.
150
HC-M, Vol. 2, p 249 and Vol. 4 (2), Ann. 23.
151
HC-M, Vol. 4(2), p 911.
152 “To understand and analyze the complex relationships between physical, chemical and biological
changes in the surface and subsurface water re gimes requires multi-disciplinary expertise in
combination with advanced mathematical modelling techniques. The overall project objective is
to establish a reliable impact assessment model for the Danubian Lowland area, which enables
the authorities to formulate optimal management strategiesleading to the protection of the water
resource and a sound ecological development for the area.” Refsgaard et al, 1994, given in HC- -48-
scientists. Professor Mucha, a Slovak groundwater specialist, was involved with the PHARE project
for a time. He wrote that: “The construction of hy droelectric power plants in this region causes new
problems for Slovakia because they affect the quality of ground water ... Many problems in this area
are as yet untouched; the answers are completely op ne... such conditions may occur which would make
groundwater unsuitable for certain purposes. The patternand rules of this complicated ecosystem is still
hidden behind a veil of mysteries.” This was again in 1993.
90. Czechoslovakia recognized the inadequacies ofthe early studies in its 1990 application to the
EC PHARE programme. The Czechoslovak Government stated that the Original Project “require[s] a
thorough and complex study of a proper impact assessment model, enabling authorities to ensure the
protection of natural and anthr op(ogen)ic resources, balanced ecological development, as well as
optimised decision making and management”. The implication of this is clear: prior to 1990 no
integrated EIA had been carried out, without such an EIA environmental protection could not be
ensured, and the appropriate management tools were not yet available. And Czechoslovakia was well
aware of these deficiencies.
91. Mr. President, Members of the Court, may I conclude this Science section by summarizing
just a few key points.
92. First of all, it is clear that large scale peak operation has ruled the entire project design, and
secondly, it is evident that impacts on the environment have not been properly addressed. By May
1989 there was no proper basis on which to determine what the impacts of the Original Project
were likely to be, or how they could be mitigated. Further studies were clearly necessary, as amply
demonstrated by the initiation of the PHARE study.
93. Many of the scientific concerns were ignored in the early studies undertaken by
Czechoslovakia and Hungary. The main thrust of the Slovak Reply, if I may say so, is simply to
deny that these risks existed. There is no attempt to demonstrate scientifically that these risks were
unfounded. And yet as I have shown, Slovak scientists, and their international colleagues, were
M, Vol. 4 (2), Ann. 12.
153 Mucha, I. (1993), “Ground water pr oblems in Slovak Danubian Lowland”
(HC-M, Vol.ume 4 (2), Ann. 11. -49-
clearly aware of them, as amply demonstrated by the annexed Slovak material, and by the internal
154
Slovak documents annexed to the Hungarian pleadings .
94. Instead, Slovakia urges the Court to turn to the grand experiment of Variant C and base its
judgement on the monitoring data. This argument is,in principle, flawed. Long-term impacts
cannot necessarily be detected within just a few years. However, as the Court will see, evidence of
Variant C, including the impact of the remedial measures proposed by Slovakia as a solution to all
problems, is beginning to demonstrate the reality of Hungary's concerns.
95. As Dr Kern and I have demonstrated, the planned operation of the Gab_íkovo-Nagymaros
Barrage System would have:
- first: endangered the quantity and quality of drinking water abstracted from
bank-filtered wells supplying two-thirds of the needs of the city of Budapest;
- second: imposed a considerable risk of pollution of valuable water resources
underneath the upper section of the Project reach;
- third: resulted in a deterioration of aquatic and terrestrial habitats, endangering
the indigenous wetland flora and fauna.
96. To return to my colleague’s original question, "how much disturbance can the Danube river
system endure without losing its essential ecological functions," the answer must unequivocally be
that the Original Project would destroy the system as we know it.
97. Mr. President, Members of the Court. That concludes this second scientific presentation on
behalf of Hungary. After the break, may I ask you to call on Ms Katherine Gorove, who will
describe the viability of the Original Project.
Thank you, Mr. President.
The PRESIDENT: Thank you, Professor Wheater. The Court will now suspend for ten
minutes. Thank you.
154E.g., SR, Vol.. II, Ann. 8; HR, Vol.. 3, Ann. 7; HC-M, Vol. 4 (2), Ann. 11; HM, Vol. 5(I), Ann. 11. -50-
The Court adjourned from 11.35 to 11.50 a.m.
The PRESIDENT: Please be seated. I call now on Ms Katherine Gorove.
Ms GOROVE:
8. THE VIABILITY OF THE ORIGINAL PROJECT
Mr. President, Members of the Court, it is aprivilege and an honour for me to address you for
the first time:
1. My colleagues have just demonstrated the grav e scientific concerns held with respect to the
Original Project. These concerns centred on: fi rst, the significant risks posed to the quality and
quantity of Hungary’s drinking water, and second, the likely negative effects on the unique species of
flora and fauna endemic to the only inland delta in Europe.
2. These concerns, and others I will mention later, impacted directly on the question of whether
or not the Original Project could proceed. But environmental considerations were not the only ones,
however important. There are other questions wh ich a reasonable government would have asked.
Did the Original Project have other substantial advantages? Or, by 1989 had the Project proceeded so
far that stopping or even substantially modifying itwas out of the question? Professor Crawford will
show tomorrow morning that the Original Project had not reached such a point of no return, that
reconsideration in 1989-1990 was still anoption. It is my task to lookat the question of whether other
benefits claimed for the Original Project were such as to provide an overriding justification for it,
notwithstanding the major risks and costs, in partic ular, those risks and costs associated with the
environment and water resources, which by 1989 were becoming clear. I will show that even under
the most hopeful of cost/benefit scenarios, and even leaving to one side the questions of
environmental risk and damage, the Project’s viability was doubtful. Under less optimistic
predictions, the Project would havenever left the drawing board.
3. Slovakia, in its pleading, states: “the Court has not been, and could not have been, asked to -51-
weigh up the economic benefits to be received by the Treaty parties and to assess their values against
155
(alleged) environmental impacts” . Slovakia further states: “the realisation or otherwise of the
expected benefits of the Project in terms of en ergy, navigation and flood c ontrol is irrelevant to
Hungary’s case and must be kept separate” .156
4. Mr. President, Members of the Court, the costsand benefits of the Project cannot be irrelevant.
Slovakia is asking this Court in 1997 to force Hungray to implement a project which was conceived in
the 1950s, designed in the 1960s, and agreed on in the 1970s. It simply cannot be the case that a
Project should be forced on a state without regard to improved knowledge or changed environmental
and financial consequences.
5. Hungary submits that a well governed stateafced with changed circumstances and about to take
irreversible steps on a large project such as thiwuld have asked itself the following three questions:
1) One, has a proper environmental impact assessment been carried out?
2) Two, if not, have the anticipated or potential consequences, both environmental and
financial, been adequately assessed?
3) Three, are the intrinsic benefits of hProject such that it should nonetheless proceed?
I will show that the answer to each of these thr ee questions, so far as the Original Project was
concerned, was no.
I.THERE WAS NO ENVIRONMENTAL IMPACT ASSESSMENT
6. I turn first to the issue of environmental impact assessment.
7. By May 1989 the use of environmental impact assessment had been accepted in principle by
Hungary and Czechoslovakia. For example, in the 1975 Helsinki Final Act of Conference and Security
and Cooperation in Europe, both countries had agreed to the use of “legal... measures for the protection
157
of the environment including procedures for establishing environmental impact assessment” . A
155
SR, para 13.28.
156
SR, para 1.48.
157 Helsinki Final Act, 1 August 1975, 14 ILM 1292 (1975), preamble. -52-
more specific commitment was made by the two countries in April 1989 with the adoption of the
158
United Nations Economic Commission for Eu rope Charter on Groundwater Management . The
Charter requires impact assessment both “at an early stage for project planning” and also “during the
construction and operative phases of a project” .9
8. What exactly did an environmental impact assessment require? At a minimum, a consideration
of potential impacts on all environmental resources,including water and biodiversity, in an integrated
manner. It must also take into account, as the Ground Water Charter specifies, an assessment of
“different alternatives”.
9. So far as the Original Project is concerned, te essential question is one of fact. As a matter of
fact, the studies carried out before 1989 did not reach even the minimum standards for an
160
environmental impact assessment. The existence o“ fhundreds of studies”, is not, as Slovakia claims ,
an alternative for a proper environmental impact assessment. In particular, Slovakia refers to the
Bioproject as a “complex’’ and “complete examination of the effect of the Project on the
161
environment” .
10. Hungary has requested access to the studies constituting the Bioproject on four occasions 162,
without success. Hungary was informed that “[t]he actulacontents of the reports are not relevant to the
contention’’ . This is curious, since the contention of Slovakia is that the Project “was indeed very
164
carefully researched’’ . As described by ProfessorWheater, the studies actually submitted by
Slovakia to the Court merely reinforce environmentlaconcerns. For its part Hungary has searched the
archives of the various officials responsible for workon the Original Project and has not been able to
158
21 April 1989, ECE/DEC/E[44].
159
Ibid, Section XIV (Impact Assessment).
160 SM, paras 2.10-2.11. SC-M, pa ras 2.17-2.22, paras 4.04-4.07.
161 SC-M, para 4.06.
162
See Note Verbale from the Republic of Hungary to the Slovak Republic, in
HC-M, Anns., Vol. 3, Ann. 14; Letters from Dr G Sz énási, Agent of the Republic of
Hungary of 6 September 1994 in HC-M, Anns., Vol. 3, Ann. 24 and Ann. 21; of 29 September 1996, in
HC-M, Vol. 3, Ann. 30; of 25 January 1995, in HR, Vol. 3, Ann. 18.
163
Letter from Dr P Tomka to G Sz énási, 3 August 1994; HC-M, Vol. 3, Ann. 11, p 38.
164 Ibid. -53-
locate the studies to which Slovakia refers.
11. Hungary has shown in its pleadings, by referen ce to the reports of various international bodies,
that there was no adequate comprehensive environmental impact assessment. 165
12. Specifically, although key concerns had b een raised throughout the 1980s by individual
scientists, the concerns had not been addressed in a comprehensive manner 166. Rather, if they were
addressed, they were addressed in isolation of other issues. For example, it might be that a study
concluded that there would be a drop in the level of ground water in the Szigetköz. Yet, that study
would stop with its conclusion. Itwould not continue with an analysis of the corresponding effects, for
example, on soils, flora, fauna, fisheries, forestry, and biodiversity.
13. The most comprehensive of the Hungarian studies on the Original Project is a 1985 study
carried out under the auspices of the Hungarian Academy of Sciences . Experts who have reviewed
and evaluated the study using international practcies for assessment have concluded that the study was
“unsatisfactory.” On a scale of “A” to “F”, they have given the study a “D”. Further, one of the two
169
reviewers points to a bias in the study . Hungarian scientists depending on state support carried out
the impact assessment.
14. As described by the professional reviewers, the 1985 study did not discuss the issues in an
integrated manner; gave no basis for the interpreta tion of the data; did not describe the standards,
assumptions or values used; and did not address the impacts of the Project on ecosystems 170.
15. As described by Professor Wheater, groups within Czechoslovakia and the Government
itself in its application to the European Communities for PHARE support, recognized the wholesale
165 Hydro-Quebec Report, HM, Vol. 5 (part 1), Ann. 9, p 298; see also HC-M, para 1.141 and HR,
para 1.72; Bechtel Report, HC-M, Vol. 4 (part 1), Ann. 1; see also HC-M, para 1.140; World
Wildlife Fund, HC-M, Vol. 4 (part 1), Ann. 4, p 5; see also HC-M, para. 1.34;
INFORT/Ecologia, HM, Vol. 5 (part 1), Ann. 5, p 59; see also HC-M, para. 1.33.
166
See HC-M, paras. 1.20-1.41 and HR, paras. 1.64-1.84.
167
See generally HR, paras. 1.74-1.75.
168
Hens, HC-M, Vol.. 4 (part 2), Ann. 1, p. 907.
169 See Hens, HC-M, Vol. 4 (part 2), Ann. 23, Ann. 1 (C), p 911.
170 Ibid, pp 888-93, 912-15. -54-
inadequacy of what was performed before 1989. In 1994, Mr.Refsgaard acknowledged the need for
advanced modelling techniques to understand and an alyse complex relationships between various
changes in water régimes in the context of di scussing Variant C and the PHARE Programme. He
stated:
“The overall project objective is to establish a reliable impact assessment model for the
Danubian Lowland area, which enables the au thorities to formulate optimal management
strategies leading to the protection of the wtaer resources and a sound ecological development
for the area."171
172
16. The Slovak PHARE Report , completed in December 1995 but made available to
Hungary only during this past month, confirms that even today further study is still necessary 173. That
Report is not itself, it should be stressed, an environmental impact assessment 174- it is the report of an
attempt to install a computer model to allow simulation of the impacts of Variant C and then only on
the Slovak side. Nor is it a substitute for an envrionmental impact assessment, which by definition has
to be done prior to the initiation, let alone the completion of a project.
17. In sum, as of May 1989, no proper environmental impact assessment had been carried out
for the Original Project in accordance with the modest standards of the time. Neither Hungary nor
Slovakia were in a position to know what then evironmental impacts of the Project would be.
171 Refsgaard et al., 1994, reprinted in HC-M, Vol. 4, Part 2, Annex 12.
172 Ministry of the Environment, Slovak Republic, and Commission of the European Communities, Danubian Lowland –
Ground Water Model, Final Report, Vol. 1 (December 1995).
173 Ibid. , Vol. 1, paras 6.2.1 and 6.2.2.
174 SR, Vol. 2, Chap. 7, para. (2). -55-
II. WERE ALL CONSEQUENCES ADEQUATELY ASSESSED?
18. I turn then to the second question, whetherin the absence of a proper environmental impact
assessment, the Project’s anticipated or potential envrionmental and financialconsequences have been
adequately assessed? As far as the environmental and scientific issues were concerned, Professor Vida,
ProfessorCarbiener, ProfessorWheater, and Dr.Kern haveshown that the answer is clearly no. It is
true that concerns have been voiced by various i ndividuals and organizations. But it was not until the
social and political changes of the late 1980s th at these concerns were heard by the Hungarian
175
Government .
19. By way of illustration, however, allow me to refer to one aspect which had not been
176
properly factored in by 1989, namely, seismicity . There is only one sentence on seismicity and large
177
reservoir impacts in the 1985 study .
20. Now it seems that there is no dispute betw een Hungary and Slovakia on two points relating
to seismicity;
- first, that the major risk was from colla pse of dykes through liquefaction – the process
whereby the ground behaves like a fluid because of intense vibrations; and
- second, that a seismic evaluation of the site should consider the “worst possible
scenario" 178. This involves, one, a determination as to the location of the closest
earthquake source to the barrages of the Original Project and, two, a determination as to
the Maximum Credible Event, the largest credible earthquake which could occur at that
site.
175 See generally HM, paras. 3.74-3.108
176 For the discussion of seismic issues in the pleadings see HM, paras. 5.99-5. 105; SM, paras. 2.12, 2.60-2.66; HC-M,
paras. 1.157-1.170; Scientific Evaluation, Vol. 2, Chap. 6; SC-M, paras. 7.105-7.114;Scientific Rebutta,l HR, Vol.
2, Chap. 8; SR, paras. 12.54-12.72;, ol. 2, Chap. 6 and Vol. 3, pp. 142-224.
177 HC-M, Vol. 4, Part 2, Annex 23, p.913.
178 See, e.g., SM, para. 2.61. -56-
21. There is disagreement, however, between Hungary and Slovakia on both of those factors. 179
Besides the physical parameters of the near surface layers 18, these twofactors are the main
components in determining an earthquake’s “peakground acceleration”. The value used for the peak
ground acceleration is determinative of whether m oderate seismic activity could result in dykes
collapsing due to liquefaction belo w their foundations. There are numerous precedents for widespread
liquefaction in ground conditions similar to thosefound in the Szigetköz under moderate shaking, as
may be expected in the “worst possible scenario” 181.
22. In fact, seismologists acknowledge the exits ence of an earthquake source zone running on a
line through Gyõr 18, a town 20 km away from Gab_íkovo (Illus No 8.1). Based on an analysis of the
1763 Komárom earthquake, about 45km away from Gab_íkovo, Slovakia presents a Maximum
Credible Event at a Richter magnitude equal to 5.7 18, and an earthquake source zone located around
Komárom 184. But estimates of the 1763 earthquake at Komárom go as high as 6.5 and indicate the
existence of an earthquake source zone much closer to Gab_íkovo than Komárom 185. Within 15-20 km
of the Gab_íkovo works, near Gyõr, small earthquakeh save been registered in the last few years 18.
23. Slovakia’s two assumptions lead it to descr ibe a 'worst possible scenario' using a peak
ground acceleration value which is not reflective of moderndesign standards. To cite an example, in
the United Kingdom, where tectonic activity is less sevee r than in the Project area, dams are designed
for four times the value used by Slovakia 187. Even using one-half the UK design standards, liquefaction
179 SR, Vol. II, pp. 89-107.
180 See Balla, HC-M, Vol. 4, Part 2, Annex 21; Bondar, HC-M, Vol. 4, Part 2, Annex 22.
181 See HC-M, Vol. II, para 6.3.3.3.
182 See Scientific Rebutta,lHR, Vol. 2, Chap. 8.1 and note 11.
183 SVR,l. III, Chap. 10.
184 SR, Vol. II, p. 98.
185 See Scientific Evaluat, HC-M, Vol. 2, Chap. 6.
186 See Scientific Rebuttal, HR, Vol. II, pp.105-106.
187 See An Engineering Guide to Seismic Risks to Dams in the United KingB,RE (1991); see contra SR, Vol. 1, para.
12.71. -57-
could typically be expected at a depth of 15m. 188. It cannot have been the case that all liquefiable
materials to this depth were removed or streng htened over the full extent of the dykes’ foundations 189.
24. Thus in 1989, Hungary had serious and we ll-founded concerns as to the integrity of the
1965 design parameters used for the Project structuresin light of developments in methods of design,
hazard evaluation, and improved safety standards. These were in addition to the major concerns
already identified by Professor Wheater, Dr. Kernn ad Professor Vida and quite apart from the sacrifice
of natural values which the Original Project entailed. Also relevant were the anticipated financial
consequences, for example, to fsiheries, forestry, and agriculture 190, and the archaeological heritage . 191
188 See, e.g., Kinitzsky, Gould & Edings, p.175 (1993).
189 See contra SR, para. 12.67.
190 Scientifvaluation , HC-M, Vol. 2, Chap. 5; Scientific Rebuttal, H,RVol. 2, Chap. 6; HM, paras. 5.68-5.98; HC-M,
paras. 1.155-1.156; 1.122-1.138.
191 HM, paras. 5.97-5.98. -58-
IV. OVERALL VIABILITY OF THE PROJECT
25. I turn then to my third question. Despite the factors that had not properly been taken into
account in planning the Project, were its benefits as of 1989 so significant that it should nonetheless
proceed? Professor Vida yesterday described why tehProject was not necessary to stop the lowering of
the riverbed and was, in fact, a primary cause of theriver’s degradation. I will turn therefore to three
192
other supposed “benefits” touted by Slovakia: (1)increased flood protection , (2)improved
193 194
navigation , and (3) energy production .
A. Flood Control
192 SM, paras. 1.21-1.34, 2.79-2.81.
193 SM, paras. 1.35-1.49, 2.82-2.83.
194 SM, paras. 1.50-1.56, 2.84. -59-
26. The first is flood control 19. The pictures of the terrible damage inflicted by the floods of
1954 and 1965 in the Slovak Memorial are evocative 19. The suggestion that the Project was necessary
for flood control in the Szigetköz region, however, isuntrue. Actually, these pictures were taken prior
to full implementation of an agreement betweenCzechoslovakia and Hungary under the 1954 Border
Waters Agreement 197. The agreement of a Government Plenipotentiary on Border Waters, which was
entirely separate to the Original Project, provide d for the implementation of design standards in
accordance with the 100-year flood 198. In other words, design standards should meet the highest flood
that may occur on average once every 100 years. Thiiss a generally accepted flood control standard for
areas such as the Project area 199. Much higher standards, such asthe flood which may occur once every
ten thousand years or once every thousandyears, are only found where decisions taken to impound a
river produce significantly higher levels of risk or where the potential damage is extraordinary 200.
27. The damage from the 1965 flood was due to the fact that 94% of the levees did not yet meet
201
these accepted safety standards for the 100 year flood . By the time of entering into the Treaty in
1977, however, the vast majority of levees along the upper reach, which includes the Szigetköz, had
been reinforced to meet the 100 year standard with adequate safety margins and the necessary
202
structures to prevent seepage had been built .
195 For the discussion of fl ood control issues in the pleadings, see Laczay, HC-M, Vol. 4, Part1, Annex 9; Scientific
Evaluation, HC-M, Vol. 2, Chap. 2.6.3; HC-M, paras. 1.172-177;Scientific Rebuttal, HR, Vol. 2, Chap. 3.2; SM,
paras. 1.21-1.34; paras.2.79-2.81; SC-M, paras. 7.118-7.121; SR, paras. 13.40-13.48; SR, Vol 2, Chap.2; SR,
Vol. 3, pp. 225-258.
196 See SM, Illus. Nos. 15, 17 (A-D)accompanying SM, paras. 1.21-1.34.
197 1954 Border Waters Agreement, 4 February 1954H,M, Vol. 3, Annex 11, 4 February 1954.
198 SLee aczay, HC-M, Vol. 4, Part 1, Annex 9, Table 2.
199 German Section of the International Commission on Irrigation and Drainage,Selection of design flood – a comparison
of international practice(in German), Schriften, Vol. 62, pp. 1-62 (1983).
200 Ibid.
201Scientific Evaluati, HC-M, Vol. 2, Chap. 2.2.4.
202Scientific Evaluati, HC-M, Vol. 2, Chap. 2.2.4; see also Lacza,yHC-M, Vol. 4, (part 1), Ann. 9, Table 3. -60-
28. Increased flood hazards, such as the potential rupture of high embankments, failure of
floodgates during high-volume discharges and the ince rased risk of flooding due to blockage of ice 203,
necessitated a 1,000 year flood level design. Because of the discharge capacity of the power canal
required for peak power operation, the dykes were at a height approximating the 10,000 year flood
204
level .
29. Thus, aspects of flood protection were in cluded in the 1977 Treaty because they were
necessary to offset the additional risks caused bythe Project itself. Apart from that, flood protection
was neither a motivating factor nor an objective justification for the Project 205. Applying a 100 year
protection level, Hungary is providing appropriate flood protection along its section of the Danube
without the Project.
B. Navigation
30. The second supposed benefit was navigation. It too was not a primary aim of the Treaty, as
Slovakia submits 20. Much as flood control, it was a by-product of the Treaty, an incidental
benefit, which could, similarly, have been achieved through other means, for example, through
traditional river training 207.
31. In the 1960s, a general river-training plan was prepared and carried out for the Rajka-Sap
208
stretch , while on the Sap-Budapest stretch, heavy dredging in lieu of river-training was carried
203Scientific Evaluati, HC-M, Vol. 2, Chap. 2.3.3.
204SM, paras. 2.79-2.81.
205For the discussion of navigation issues in the pleadings see HM, paras. 5.91; Scientific Evaluation, HC-M, Vol. 2, Chapter
2.6.3; HC-M, paras. 1.178-1.189;Scientific Rebutta,lHR, Vol. 2, Chapter 3.1; SM, parsa. 1.35-1.49; paras. 2.82-2.83; SC-M,
paras. 7.115-7.117; SR, para. 3.31-13.39; SR, Vol. 2, Chap. 2; SR, Vol. 3, pp. 263-265.
206SM, para. 1.46.
207See contra SM, paras. 1.39-1.40, 1.45.
208Scientific Evaluati, HC-M, vol. 2, Chap. 2.2.3. -61-
209
out in anticipation of the Project . Therefore, as of 1977, upstream there were few obstacles to
navigation, while downstream was more problematic because of years of neglect.
32. Since the termination of the Treaty, an international expert group has examined various
options for improving navigation: one of its recommendations using a combination of traditional
river training and dredging would achieve a fairway depth of 2.7 metres and a width of
210
120 metres .
33. Slovakia claims that in 1977 "the stretch of the river between Bratislava and Budapest was
navigable on average only 120 days per year" 211. This is entirely misleading. For the vessels
which actually use it, the Danube was and is generally navigable in this sector throughout the year,
except in situations of blockage by ice, which the Original Project would not have improved.
34. The term "navigable" is being used by Slovakia to mean navigation with conditions
corresponding to the recommendations of the Danube Commission. Slovakia is giving more
weight to these recommendations than they were intended to bear, as these are non-binding
212
recommendations passed by majority vote of Danube Commission members .
35. Is a channel with dimensions different from those recommended by the Danube
Commission navigable? Certainly yes. Three points should be made:
(1) First, the Danube Commission's recommendations vary according to
209See Laczay, HC-M, Vol. 4 (part 1), Ann. 8.
210See Delft-Harris-VITUKI, Danube Environmental and Naviga tion Project, Feasibility Study , Rajka-Budapest, Final
Report, Stretch B1: Szap-Ipoly Mouth (August 1994), placed on file with the Library of the Court. See also Delft-Harris-
VITUKI, Danube Environmental and Navigation Project, Feasibility Study , Rajka-Budapest, Final Report, Analysis of
Strategies, Stretch A: IpolyMouth - Budapest (December 1993).
211SR, para. 13.32.
212Convention concerning the Regime of Navigation on theDanube, Belgrade, on 18 August1948, 33 UNTS 181, Art. 11,
reprinted in HM, Vol. 3, Ann. 4. -62-
each stretch of the River. Depending on the class of ships, navigability
may either be more or less difficult on certain stretches. In the
Gab_íkovo-Nagymaros stretch the recommended width is 100-180 metres
across, with a ship draught of 2.5 metres most of the year in the
213
unimpounded stretch . In other stretches of the Danube, the
recommendations may be far less, and on other rivers the standards far
less. For example, on the Main the width is generally 40 metres or
less.214 Where there is difficulty with ship navigability due to narrow
stretches, the accepted practice is to have ship traffic regulations 215.
(2) Second, contrary to Slovak claims, the Nagymaros reach is not the worst
bottleneck on the Danube River. The Danube between Straubing and
216
Vilshofen has a water depth of 2.5 metres only 150 days per year and is
acknowledged as the worst bottleneck on the fairway between the Black
Sea and the North Sea 21.
3) Third, the importance of river navigation on the Danube has declined.
Even if the Project had increased traffic, it would have been relatively
insignificant, especially when compared to the amount of traffic on rivers
such as the Rhine, where there is one ship after another. On the much
broader Danube, there is far less traffic. In fact, the volume of
merchandise transported on the Danube has dropped by about one-half
213Danube Commission RecommendationsS , M, Vol. 2, Ann. 14.
214Network of Inland Waterways(in German), (Verein fuer Biunenschifafahrt und WassenstraBen VBW, editors), 1991.
215See H Contzen, Remarks on the Regulation of the Danube between Straubing and Vilshefen(in German), Zeitschrift fuer
Binnenschiffahrt 18, Setmber 1993; D Eujen,Regulation of the Fairways of theMain between Aschaffenburg and Bamber,g
Zeitschrift fuer Binnenschiffahrt, 23/24, December 1993.
216See Ibid; p. 12, question 35; see also H Contzen,Remarks on the Regulation of the Danube between Straubing and
Vilshefen(in German), Zeitschrift fuerBinnenschiffahrt 23/24, December 1993.
217Answer of the Government of the Federal Re public of Germany to Queries Posed by Members of Parliament, 11/2/93,
Publication No 12/4351 (in Geran), p. 13, Question 39. -63-
between 1980 and 1991 218. Further, the relative importance of river
navigation as opposed to rail or road has sharply declined.
36. In conclusion, in terms of navigation, the Project would have addressed concerns which
could have been met through traditional river training. It cannot be argued that the Project's
219
improvements to navigation contributed significantly to the viability of the Project .
C. Energy Production
220
37. I turn next to the third supposed benefit, energy production . Slovakia argues that "the
221
'value' of [power generation] does not touch on the questions before the Court in this case" . At
the same time, it claims that: Hungary's reasons for ceasing its investment into the Project had an
222
economic rather than environmental basis" .
38. In fact energy production was intended to be the primary benefit of the Original Project, as
well as the primary mechanism by which the Project would have been financed 223. The value of the
224
power to be generated is critical to a determination of viability .
39. The relative importance of the electric energy to be produced by the Project has steadily
declined. When the Project was initially conceived in the 1950s, Hungary's share of its projected
output was approximately equivalent to 30% of its needs. By the 1960s, when project planning
began in earnest, projected output would have accounted for some 20% of its needs. When the
Project was being agreed upon within the framework of the 1977 Treaty, projected output was no
more than 8.5% of Hungarian electric energy needs 225. At present levels of demand, only 5% of
218Annuaire Statistique de la Commission du Danubeour 1992 (1994).
219See Equipe Cousteau, Final Report, The Danube... For Whom and for What? The Gab_íkovo Dam: a Textbook Case
(March 1993), reprinted in HM, Vol. (Part 2), Ann. 16, pp. 567-568.
220For the discussion of energy issues in the pleadings, see HC-M, paras 1. 190-1.203; SM, paras 1.50-1.56 and 2.84; SR,
paras. 13.26-13.30; and , Vol. 3, pp. 259-262.
221SR, para. 13.28.
222SR, para. 13.27
223SR, para. 13.30.
224See HC-M, paras. 1.190-1.203; see also Equipe Coustea, he Danube... For Whom and for What? The Gab_íkovo Dam: a
Textbook Case(September 1992), reprinted in HM,Vol. 5 (Part 1), Ann. 12, at 354-361.
225HR, para. 1.199. -64-
Hungarian consumption would be met 226. According to 1994 data supplied by the International
Energy Agency, the Czechoslovak share would have been less than 5% of total consumption of the
227
Czech and Slovak Republics .
40. Furthermore, since the political changes in 1989, most governments have implemented
conservation-minded energy policies and industry in the region has generally become more energy
efficient. The days of huge energy subsidies in order to support artificial production goals without
regard for economic, environmental or social effects are over 22. Recent predictions indicate that
through the year 2000, the per capita demand for energy will continue to decrease in Central and
229
Eastern Europe . As can be seen from the diagram on the screen, energy consumption in
230
Hungary has returned to approximately the level of 1983 (Illus. No. 8.2) .
41. And what of the net value of the electric energy produced by the Project? With little by
way of supporting documentation, Slovakia asserts that Variant C currently nets in excess of US$
100 million per annum 231. There is not enough information to determine if all costs are adequately
accounted for and income stream assessments made according to generally accepted accounting
procedures. For example, it would be interesting to know whether Hungary's capital contributions
to structures and equipment are calculated as an asset or a liability or, at the costs of remedial
measures been factored in. Slovakia has so far denied the Court the tools necessary to assess the
validity of this claim.
226HR, para. 1.192.
227International Energy Agency,Energy Statistics and Balances of Non-OECD Countrs993-1994 (1996).
228See contraSR, para. 13.28.
229International Energy Agency, World Energy Outlook (1996).
230International Energy Agency,Energy Statistics and Balances of Non-OECD Countrs993-1994 (1996).
231SR, para. 13.29. -65-
D.Economic Analyses
42. I now turn to a description of what the economic analyses indicate about the overall viability
of the Project. By 1989, Hungary had sufficient cause to question the viability of the Project. In fact,
the economic analyses carried out in Hungary, alh tough themselves weighted in favour of carrying out
the Project, actually demonstrated its marginal value.232
43. In particular, all calculations made prior to 1989 indicated that costs exceeded benefits. For
233
such analyses to reach favourable conclusions on the Project’s viability as they did, indirect
economic benefits were given undue weight, while various costs were ignored. In particular
environmental costs were inadequately assessed to the point that they were virtually ignored. Other
costs that would have to be expended to maximise aparticular “benefit” were also neglected – such as
the costs of relocating a port to improve navigation.
44. The economic analyses of the Project done in Hungary also ignored what was happening
outside of Central and Eastern Europe. The underl ying assumptions were that the future would be
similar to the past. For example, there was little or no analysis of alternative ways of producing the
same amount of energy, the costs of the capital itself orthe costs of conserving energy versus the costs
of expanding supply were not takeninto account. InfIationary price changes were not considered, and
no account was taken of relative currency values. 234The Joint Contractual Plan 235appears to confuse
the term “profitability” with the term “revenue.” Perhaps this is why the European Bank for
232
See Norgaard, The Economic Analyses of theGabcikovo-Nagymaros Barrage System: A Report,
HR, vol 2, app 4.
233
See, e.g., István Varga, The Dynamic Analysis of the GNBS, February 13, 1985; Resolution No
3540/1975 of the Ministerial Council on the Investment Proposal for the GNBS, November 20,
1975; The Economic Efficiency Study of the GNBS, February 9, 1983. National Office of Water
Management, Modified Investment Proposal for the GNBS, State Investment and Evaluation of
its Technical-Ecological-Economic Aspects, February 1986 and July 1986. National Planning
Office, Feasibility Calculations of the Gabc ikovo-Nagymaros System of Barrages , October,
1989. A copy of all documents ha ve been deposited with the Court.
234 See generally Norgaard, HR, vol 2, app 4.
235
Joint Contractual Plan, Summarizing Documentato in, 0-6 Economic Part, June 16, 1978. A copy
of the document has been depositedwith the Court, items 7 and 13. -66-
236
Reconstruction and Developmentreferred to the Project as ben ig of “dubious economic value.”
45. A central issue in the transformation in Ce ntral and Eastern Europe was the change in
perception of the relations betweeneconomic development and the environment. The change greatly
affected economic analyses of projects and their viability. Pre-transition analyses assumed that the
numerous environmental complications could be corrected without any additional cost.237
46. In the words of Professor Norgaard...
“there are many projects proposed which do not receive full economic evaluation because
professional economists judgea priori that these projects are not economically viable....[The
Original Project] falls in this category of projects. If a project similar to the [Original Project]
were pr238sed today...it would not receive a full evaluation and would probably be rejecteda
priori.”
47. In summary, by May 1989 Hungary had good reasonsto believe that the original Project was
not economically viable, all the more so when one took into account the significant risks to natural
resources and environmental and financial costs.
236
Letter from T Baudon, Director, Infrastructure, Energy & Environmental Department, European
Bank for Reconstruction and Development, to Messrs C Balint & P A Farkas, Reflex, 19 May
1992.
237
IstvánVarga, The Dynamic Analysis of the GNBS, February 13, 1985. A copy of the document
has been deposited with the Court.
238 See Norgaard, HR, vol 2, app 4, pp 179-180. -67-
IV. C ONCLUSION
48. To conclude, Slovakia has argued that financia l considerations were the sole motivation for
Hungarian actions. 239In fact, the financial analyses done before 1989, although using calculations
which indicated that costs were greater than benefits, always came to a favourable conclusion on the
viability of the Project. It was only in the light of the dramatically changing political and economic
situation of the late 1980s that these analyses could be revisited, at a time when at last the values
associated with protection of the environment and drinking water resources were also increasingly
taken into account. Some of the ancillary benefitsof the Project, such as improvements to navigation
were real, but none of them were of such significance so as to justify the Project as a whole. In the
circumstances, and taking into account the inadequaceis of the environmental impact analyses done to
that point, Hungary acted reasonably in calling for reconsideration of the Project, and in suspending
further steps until that reconsideration could take place.
49. Mr. President, Members of the Court, a reasonable state faced with significant questions as to
Project viability would have called for a joint comrpehensive environmental impact assessment so as to
discover all the likely consequences before proceenig. But Slovakia claims that once the 1977 Treaty
was concluded, no subsequent developments, no matter how pressing the necessity, could justify
departure from that Treaty. I ask you now to call on Professor Dupuy to examine that claim.
The PRESIDENT: Thank you so much, Ms Gorove. Professor Dupuy.
M. DUPUY :
9. LES RÈGLES DE DROIT JUSTIFIANT LA SUSPENSION ET LA TERMINAISON DES
TRAVAUX PAR LA HONGRIE
239 SM, paras 3.36, 3.56, 3.6. -68-
1. Messieurs les juges, c’est pour moi un honneur et un plaisir renouvelé de me présenter à
nouveau devant vous à cette barre. Ma tâche, tout au long de ces plaidoiries, consistera dans
l’exposé d’un certain nombre de règles et d’arguments de droit applicables en la présente affaire.
Aujourd’hui, il s’agit pour moi de vous exposer le fondement juridique des raisons pour lesquelles
la Hongrie s’est vue amenée à suspendre puis à terminer les travaux sur certains sites à partir
de 1989. Mon collègue James Crawford reviendra en détails sur ces faits demain matin. Je n’en
dirai moi-même donc que très peu de choses, destinées, à titre liminaire, à vous rappeler pourquoi,
en cette affaire, l’invocation de l’état de nécessité occupera une place accrue, entre le printemps de
1989 et celui de 1992, entre l’interruption des travaux et la terminaison du traité.
La divergence croissante de chacune des deux Parties à l’égard de la rationalité du projet de
240
barrage devait en effet prendre un tour dramatique à partir de mars 1989 . Plusieurs éléments
expliquent ce phénomène. En premier lieu, la confirmation définitive des préoccupations de la
241
Hongrie . Elle était jusque là inquiète; elle devient extrêmement alarmée quant aux risques très
considérables engendrés par la réalisation du projet pour l’équilibre écologique de toute la région
concernée. Tous les experts, hongrois et étrangers, insistent en particulier sur l’importance du
danger créé pour les ressources en eau potable. La Hongrie estime qu’il faut prendre le temps
d’évaluer la situation et réfléchir sur les conséquences de droit et de fait à en tirer, afin de
renégocier une partie du traité de 1977. Elle décide alors l’interruption des travaux à Nagymaros et
elle offre en même temps à l’autre partie d’ouvrir des négociations.
2. C’est à partir de ce moment-là qu’on assiste à un véritable dialogue de sourds 242: la
Hongrie invoque la responsabilité des deux gouvernements envers les générations présentes et
243
futures . La Tchécoslovaquie répond en invoquant ses besoins en énergie et la possibilité de
réduire les dommages à l’environnement par de simples aménagements techniques. Jamais, elle
240
Voir § 9.04 et suiv. et §§ 3.74 à 3.108 du mémoire hongrois.
24Voir R.H. § 3.101.
242
Voir MH, § 9.06 et suiv.
24Cf. MH, § 3.78 et 3.93. -69-
244
n’envisagea de remettre en cause ne fût-ce qu’une partie du projet initial . Dans un tel contexte,
l’édification de la variante C par la Tchécoslovaquie entre 1990 et 1992 paraît inexorable. Les
efforts constamment renouvelés de la Hongrie pour provoquer le réexamen substantiel du traité
resteront sans succès. Aussi la concrétisation progressive de la menace de dérivation unilatérale du
Danube constituera, durant toute cette période et c'est là le point important un facteur majeur et
constant d’aggravation de la situation à l’origine de ce différend...
On comprend dès lors pourquoi, à la fin de cette période, la déclaration du 16 mai 1992
présentant les raisons de la terminaison du traité de 1977 devait commencer par le rappel de
l’argument de nécessité déjà invoqué en 1989 245.
3. Il résulte des éléments de fait qui précèdent qu’en droit, l’invocation de l’état de nécessité
joue effectivement dans cette affaire un rôle considérable, quoique, évidement, non exclusif. Nous
examinerons d’abord les conditions juridiques mises par le droit international à l’invocation de
cette circonstance exceptionnelle (I). Pour autant, on ne saurait s’en tenir là. La Slovaquie, en effet,
dénie à la Hongrie le droit d’invoquer toute circonstance excluant l’illicite, quelle qu’elle soit. Et
elle le fait au motif que, s’agissant d’un traité, la suspension de sa mise en oeuvre et, à fortiori sa
terminaison, ne relèveraient ... que du droit des traités. La Hongrie serait ainsi interdite
d’invocation de tout motif d’exclusion de l’illicite, puisque ces derniers relèvent du droit
international de la responsabilité. Il conviendra alors de rappeler l’inanité d’une telle thèse, et tout
d'abord la justification de l'invocation de l'état de nécessité (II).
I. JUSTIFICATION DE L ’INVOCATION DE L ’ÉTAT DE NÉCESSITÉ
4. L’état de nécessité, en droit international comme ailleurs, est une circonstance
exonératoire de responsabilité. Il se distingue pourtant totalement des autres causes exonératoires,
que la CDI appelle des «causes excluant l’illicite». L’état de nécessité, ce n’est ni la force majeure
ni le cas fortuit, parce que ces deux autres cas exonératoires sont des cas dans lesquels la volonté
de l’auteur de l’acte est pratiquement absente, comme c’est par exemple le cas d’une catastrophe
24Cf. MH, § 3.84.
245
Cf. MH, § 10.03 et suiv. -70-
naturelle. L’état de nécessité, au contraire, comme le dit à juste titre Roberto Ago “implique un
246
comportement librement et volontairement adopté” .
Pour autant, l’état de nécessité se distingue également des contre-mesures, avec lesquelles il
partage pourtant ce caractère volontaire. Contrairement à elle en effet, cette cause exonératoire peut
être être invoquée même sans constituer une réaction à un fait illicite initial commis par l’autre
Etat. C’est d’ailleurs la raison pour laquelle le pays qui invoque à juste titre l’état de nécessité
devra le plus souvent acquitter une compensation, généralement sous forme d’indemnité, à son
partenaire affecté par la non-réalisation de son obligation, ici, une obligation contractuelle.
En l’espèce, les circonstances ayant justifié la conduite hongroise étaient bel et bien
provoquées par le comportement de la Tchécoslovaquie. Celle-ci avait en effet méconnu, au
minimum, les clauses du traité de 1977 concernant la protection de la nature (articles 15,19 et 20)
et, surtout, celles qui impliquaient la coopération et la constante disposition à négocier
véritablement. Cette attitude impavide a certainement contribué à la formation d’une situation
dangereuse. Pour autant, j'y insiste, même si la révélation des risques majeurs inhérents au traité
n’avait rien dû à l’attitude de la Tchécoslovaquie, l’état de nécessité aurait néanmoins pu être
invocable par la Hongrie.
C’est que l’état de nécessité vise avant tout une situation, et non d’abord un comportement
du partenaire. Cet argument de nécessité présente en tout état de cause un caractère exceptionnel.
L’admettre d’une manière trop peu rigoureuse entraînerait bien évidemment des menaces très
sérieuses pour la sécurité des rapports juridiques.
5. Messieurs de la Cour, cela revient à dire que la question de savoir si un état de nécessité
existait ou non dans une situation considérée au moment où il fut invoqué dépend dans une très
large mesure de l’analyse de ces circonstances de fait sur lesquelles reviendra demain mon ami
James Crawford. C’est une question qui, par excellence, s’analyse in concreto. Alors cette analyse
peut résulter de l’examen conjoint de la situation par les parties intéressées lors d’une négociation.
Cependant, quand le refus systématique de l’une d’entre elles de reprendre au fond l’examen de la
246
R.Ago, Annuaire de la CDI 1980, vol. I, p. 144, §39. -71-
situation est avéré, il est naturel que ce soit une tierce partie qui apprécie s’il y avait ou non en
l’occurrence “état de nécessité”. Il est d'ailleurs difficile de contester que la Hongrie ait jamais dit
autre chose : lorsqu’elle a constaté qu’elle ne parviendrait pas à convaincre la Tchécoslovaquie de
renégocier le traité qu'a-t-elle fait ? Elle l'a invitée à venir vous trouver; cela déjà en date du 18
août 1992 247.
6. C’est aussi parce ce que cet argument de nécessité est éminemment “justiciable”, c’est-à-
dire susceptible d’appréciation par le juge, que la Hongrie a été amenée, dès le stade de son
mémoire, à produire l’ensemble des expertises scientifiques et techniques qu’elle a commanditées.
Ces expertises, j'en conviens, peuvent nous paraître austères à nous juristes, elles sont pourtant
incontournables, parce qu’elles vérifient, dans toute la mesure où la rigueur scientifique le permet,
que les profondes préoccupations de la Hongrie dès le milieu des années quatre-vingt étaient
justifiées.
7. En droit, la question est simple : dès le stade de l’interruption des travaux, et, à fortiori
trois ans plus tard, la Hongrie pouvait-elle s’estimer légitimement fondée, à considérer que
l’ampleur des incertitudes engendrées par la construction des barrages obligeait tout gouvernement
raisonnablement diligent à interrompre, puis faute mieux, à terminer ces travaux ?
8. Etant donné le caractère exceptionnel de cet argument de nécessité, on comprend que la
Commission du droit international, qui l'a codifiée, ait adopté à l’égard de cette cause exonératoire
une position équilibrée.
D’un coté, elle ne voulait pas soumettre un Etat à des contraintes allant à l’encontre du
devoir général qu’il a de protéger les intérêts majeurs de son territoire et de sa population. De
l'autre, il fallait bien qu'elle restreigne les conditions d'invocation d'une telle exonération. Alors elle
a pris un parti justifié. Il consista à rédiger l’article 33 du projet de codification du droit de la
responsabilité dans une forme négative. Il dit : “l’état de nécessité ne peut pas être invoqué par un
Etat comme une cause d’exclusion de l’illicéité d’un fait de cet Etat non conforme à une de ses
obligations internationales”. Cela, c’est la règle générale.
247
Cf. MH. vol. 4, annexe 92, p. 197 et vol. 1, §§ 3.175 et suiv. -72-
Puis il énonce les exceptions à cette règle, regroupées autour de deux conditions
cumulatives. Il faut en premier lieu que ce fait “ait constitué le seul moyen de sauvegarder un
intérêt essentiel dudit Etat contre un péril grave et imminent”. Il faut, en second lieu, que ce fait
“n’ait pas porté atteinte à un intérêt essentiel de l’Etat à l’égard duquel l’obligation existait”.
9. Ainsi que le dit M. Ago, “La nécessité est une situation de fait dans laquelle un Etat, lié
envers un autre Etat par une obligation internationale, se refuse à exécuter cette obligation, car par
là il porterait atteinte à l’un de ses intérêts essentiels» 248.
Qu’est-ce qu’un intérêt essentiel ? Comme l’indiquait le rapporteur spécial, ce n’est pas
249
nécessairement “un intérêt à l’existence” . Ce sont des intérêts qui, selon lui, “peuvent relever de
domaines aussi divers que l’économie ou l’écologie”. A sa suite, au demeurant, la Commission du
droit international voulut indiquer que le caractère “essentiel” de l’intérêt en cause peut en
particulier concerner la nécessité de garantir la survie de la faune et de la flore de certaines régions
250
ou, plus largement, nous dit-elle, celle de préserver l’équilibre écologique de toute une région .
Le rapporteur spécial s’était référé à bien des cas dans lesquels l’argument de nécessité a été
invoqué avec succès, comme l’illustrent notamment l’affaire du “Neptune”, celle de la “Caroline”
251
ou bien encore celle relative aux droits des ressortissants des Etats-Unis au Maroc . Mais il a
évoqué également un autre précédent, particulièrement significatif de l’ancienneté des
préoccupations de sauvegarde de la nature.
Il s'agissait non pas de l'affaire des forêts du Rhodope, malgré son nom bucolique et le fait
que l'argument de nécessité y avait été accepté mais de celle des "Pêcheries d'otaries à fourrure au
large des côtes russes". Elle était écologique avant la lettre cette affaire puisqu'elle remonte à la fin
e
du XIX siècle. On sait que le Gouvernement britannique y admit l'invocation d'un Etat de
nécessité par la Russie, lorsque celle-ci avait dû se résigner à interdire la chasse aux otaries alors
24Annuaire de la CDI 1980, vol.I , p. 144, § 41.
24Ibid., p. 146, § 6.
250
Voir MH, p. 285, § 10.10.
25Ibid., p. 149. -73-
même que les lieux de chasse se situaient en dehors des zones placées sous sa juridiction. Or, dans
notre affaire, la Hongrie n’a pas invoqué la nécessité pour sauver un troupeau d'otaries. Elle l'a
certes fait pour sauver la flore et la faune de la zone humide du Szigetköz. Elle l’a fait également
pour sauvegarder la beauté du site exceptionnel de Nagymaros. Mais elle l'a fait, surtout, pour
sauvegarder l'intérêt des générations présentes et, plus encore, futures, dans la garantie de leur
approvisionnement en eau potable.
Qu’est-ce que cela, sinon ce qu'on peut appeler en droit un «intérêt essentiel» ?
Elle l'a fait aussi par référence à l'appréciation comparée des coûts économiques, des aléas
écologiques et des très contestables bénéfices énergétiques de l'entreprise. C'est l'ensemble de ces
éléments qui constituait, dès 1989, cet «état de nécessité écologique» dont la Slovaquie, à défaut de
pouvoir vraiment contester la notion, met aujourd'hui en cause le contenu.
10. Il ne suffit cependant pas que l'intérêt en cause soit essentiel. Il faut aussi que le péril à
conjurer soit imminent. A cet égard, il ne m'appartient pas de revenir sur les circonstances de
l'espèce mais je me contenterai de rappeler que c'est l'attitude tchécoslovaque en faveur du maintien
intransigeant de l'intégrité du traité qui a d'abord placé la Hongrie devant une situation dans
laquelle elle n'avait d'autre issue, pour tenter d'arrêter le temps, que d'interrompre les travaux.
11. Il en va de même pour la dernière condition que doit remplir l'invocation de l'état de
nécessité. Elle doit se rapporter au caractère inévitable de la décision. La seule autre façon de
parvenir au but poursuivi par la Hongrie, c'était la renégociation. Cette renégociation, la
Tchécoslovaquie l'a assez dit, elle n'en voulait pas. Elle voulait bien des aménagements techniques
mineurs, mais pas des revisions substantielles.
Alors, Messieurs les juges, ce sera à vous de répondre : quel autre moyen restait-il à la
Hongrie pour tenter de sauvegarder ses intérêts essentiels, et, au demeurant, non seulement les siens
mais ceux d'une région dépassant ses propres frontières.
Pour le reste, à moins que la Slovaquie n'arrive à vous démontrer que le traité de 1977
incorporait une ou plusieurs règles de droit impératif auxquelles l'invocation de l'état de nécessité -74-
par la Hongrie aurait porté atteinte, je ne vois pas, pour ma part, laquelle des conditions énoncées
par l'article 33 du projet de la CDI n'aurait pas été, en l'occurrence remplie.
C'est alors, parce qu'elle est consciente du sérieux de la position hongroise, que la Slovaquie
n'avait que deux moyens d'y faire front : soit contester la bonne foi hongroise, soit tenter de faire
barrage, c'est le cas de le dire, à l'argument de nécessité en disant qu'il n'est pas invocable parce que
l'obligation à laquelle la Hongrie n'a pas déféré était, en l'espèce, d'origine conventionnelle.
II. Réfutation de la thèse slovaque de non invocabilité du droit international de la responsabilité par
la Hongrie
12. L'article 2 du compromis sur la base duquel la Cour internationale de Justice se trouve
saisie du présent différend lui pose des questions simples. La Hongrie était-elle fondée à
interrompre puis à abandonner ses travaux dans le cadre du Projet relatif au système de
252
barrages ? La Tchécoslovaquie avait-elle le droit de mettre en Œuvre la Variante C ? Si la
Hongrie n'avait pas le droit d'agir comme elle l'a fait, elle engage sa responsabilité à l'égard de la
Slovaquie et la réciproque est vraie dans le cas symétrique où les agissements de la
Tchécoslovaquie puis de la Slovaquie seraient illicites. Rien ici que de très classique. Comme l'a
rappelé constamment la juridiction internationale, «c'est un principe de droit international que la
violation d'un engagement entraîne l'obligation de réparer» 253.
Pourtant, l'étrange position que la Slovaquie tente de défendre est de prétendre que, pour
trancher ces questions de responsabilité internationale, la Cour ne pourrait pas appliquer ... le droit
de la responsabilité internationale !
D'après la Slovaquie, la Cour ne le peut pas parce que l'obligation ou les obligations
susceptibles d'avoir été violées trouvent leur fondement dans un traité. Il faudrait donc, d'après
elle, appliquer le droit international des traités et lui seul, parce qu'il comporte lui-même des règles
relatives à la suspension et à la terminaison des traités. Et ces règles, prétend la Slovaquie, seraient
25Article 2. 1
25Usine de Chorz¢w, C.P.J.I. sérieA n 9, p. 21. -75-
exclusives de celles du droit de la responsabilité internationale 254 !
Cette opinion s'appuie sur le présupposé que le souci de garantir la stabilité des conventions
a incité les auteurs de la convention de Vienne de 1969 sur le droit des traités à restreindre pour un
Etat le droit de suspendre ou de mettre fin à un traité au seul cas de «violation substantielle» du
traité par l'autre partie. Cette expression est, comme on sait, définie à l'article 60 de la convention,
pôle de tout le raisonnement juridique slovaque.
13. La Cour perçoit ainsi d'emblée ce qui lui est demandé par la Slovaquie. Elle lui
demande, ni plus ni moins, de faire de l'arrêt qu'elle va rendre en la présente affaire un arrêt de
principe ! Un arrêt, qui, à n'en pas douter, fera date dans l'histoire de la jurisprudence
internationale par sa nouveauté ... radicale.
Il s’agirait en effet d'affirmer en droit international l'existence d'un double régime de
responsabilité. A l'imitation de certains droits internes, on distinguerait ainsi désormais une
responsabilité délictuelle ou quasi délictuelle, lorsque les faits incriminés méconnaissent une
obligation trouvant son origine hors d'une convention; et puis d'autre part, une responsabilité
«contractuelle» ou «conventionnelle» lorsque les obligations méconnues sont tirées d'un traité,
comme c'est ici le cas. Le droit de la responsabilité se trouverait, du même coup, dépossédé d'une
très large part de son champ d'application, quand on sait la part prise aujourd'hui par la création
d'obligations internationales sur une base conventionnelle.
Un arrêt de la Cour faisant droit aux requêtes slovaques aurait enfin une portée d'autant plus
considérable qu'il irait évidemment à l'encontre de la règle codifiée à l'article 17 du projet de la
commission sur le droit de la responsabilité dont toute la doctrine s'est pourtant accordée à
reconnaître qu'il traduisait très fidèlement le droit international coutumier.
Dois-je le relire ?
«1. Le fait d'un Etat qui constitue une violation d'une obligation internationale
est un fait internationalement illicite quelle que soit l'origine, coutumière,
conventionnelle ou autre, de cette obligation.»
254
MS, par. 8.13. -76-
Et la même disposition précise à son paragraphe second :
«2. L'origine de l'obligation internationale violée par un Etat est sans effet sur
la responsabilité internationale engagée par le fait internationalement illicite de cet
Etat.»
14. Aujourd'hui, au contraire, si l'on suit la thèse slovaque, il faudrait distinguer d'après
l'origine de l'obligation non respectée.
15. Un arrêt de votre part consacrant la thèse slovaque, Messieurs les juges, serait d'autant
plus assuré de passer à la postérité qu'il irait également à l'encontre de l'ensemble de la
jurisprudence qui vous précède. Votre arrêt s'opposerait notamment à la dernière espèce rendue en
la matière, celle du Rainbow Warrior, dans laquelle la Nouvelle-Zélande soutenait exactement la
même thèse que la Slovaquie d'aujourd'hui et la France une position identique à celle de la Hongrie
dans notre affaire. Or, en application des règles classiques, ce fut, bel et bien, sur ce point tout au
moins, la France qui l'emporta.
De manière, en effet, parfaitement orthodoxe, le tribunal présidé par, l'ancien président de
cette Cour, Monsieur Eduardo Jiménez de Aréchaga déclara :
“les conséquences juridiques de la violation d’un traité, y compris la détermination des
circonstances aptes à exclure l’illicéité et la réparation appropriée en cas de violation
sont des questions de droit coutumier [portant] sur la responsabilité des Etats”.
La sentence ajoutait:
“la raison en est que les principes généraux du droit international en matière de
responsabilité des Etats sont également applicables en cas de manquement à une
obligation d’un traité puisqu’en droit international, on ne fait pas la distinction entre
responsabilité contractuelle et responsabilité pour acte illicite.” 255
Il est vrai qu'un désavŒu évidemment parfaitement concevable de cette jurisprudence arbitrale par
vous-même satisferait les exigences de la Slovaquie qui nous déclarait :
“Slovakia contends that this arbitral award does not correctly state the
relationship between the law of treaties and the law of State responsibility, and
reserves its right to invite the Court so to find, in the context of the dispute between
Slovakia and Hungary.”
Voici donc la Cour avertie. Elle se trouve investie par la Slovaquie d’une compétence d’appel (ou
255
Texte reproduit dans Revue générale de droit international public, 1990/3, p. 851, par. 75. -77-
de cassation?) à l’égard d’une sentence arbitrale rendue entre deux Etats tiers et pourtant
unanimement approuvée, en tous cas sur ce point!
16. Face à l’argument slovaque d’exclusion du droit de la responsabilité par celui des traités
quant aux conséquences de la violation d’une obligation contractuelle, le contre-mémoire hongrois
avait fait bien sûr quelques objections. Il s’était notamment permis d’indiquer en substance que si
ces deux droits, l’un et l’autre profondément enracinés dans le droit coutumier international, étaient
exclusifs l’un de l’autre, il était surprenant que l’on ait attendu la plaidoirie de la Slovaquie, faite à
e
l’extrême fin du XX siècle, pour finir par s’en rendre compte!
Apparemment ébranlée par ce simple constat de bon sens, la République slovaque a d’abord
semblé faire certaines concessions 256, mais en réalité elle continue à nier la réalité du droit positif.
Alors qu'elle est-elle cette réalité ? Elle est toute simple. Elle est que le droit des traités et le droit
de la responsabilité, selon la Hongrie, ont chacun un domaine d’application distinct. Ceci permet
d’expliquer la précaution bien connue prise par la convention de Vienne sur les droits des traités, à
son article 73, lequel précise que ses dispositions “ne préjugent aucune question (je souligne,
aucune) qui pourrait se poser ... en raison de la responsabilité internationale d’un Etat”.
17. Malgré tout, dans sa réplique, la Slovaquie continue à traiter du droit des traités et du
droit de la responsabilité comme s’ils étaient placés en position de concurrence, de rivalité ou
d’exclusion l’un par rapport à l’autre. Sa thèse continue à être, alors même que la Hongrie n’a
jamais rien affirmé de semblable, que le droit de la responsabilité n’ajoute pas de nouvelles causes
de suspension, d’exécution ou de terminaison à celles qui sont prévues dans l’article 60.
Elle persiste et signe dans ses errements passés quand elle affirme que l’article 73 ne dit pas
ce qu’au demeurant personne ne lui fait dire. 257 Et puis un peu plus loin, elle s'engage dans une
construction intellectuelle, à vrai dire assez complexe, de laquelle il ressort que si l’on veut
chercher une complémentarité du droit des traités et du droit de la responsabilité, il faudrait bien
256
RS, p. 86, par. 4.14.
25“The ILC’s dictum does not refer to additional termination grounds being a ma tter failing within State responsibility”
(RS, par. 4.20). -78-
peut être la chercher du côté de l’article 60 de la convention de Vienne sur le droit des traités lequel
ne s’occupe que des cas de suspension ou de terminaison des traités pour cause de violation
substantielle, cependant que les violations mineures relèveraient du droit de la responsabilité, réduit
au rôle, ô combien ingrat de “ramasse miettes”. Curieuse ligne de partage : au droit des traités
reviendrait la gestion des violations substantielles; au droit de la responsabilité resterait celle des
violations mineures! Voilà bien une conclusion insolite, d’ailleurs formulée avec une certaine
gêne, je cite la Slovaquie : “It may be that the law of State responsibility will have a role to play
258
here” .
18. Relisons alors simplement ensemble, si vous le voulez bien, Messieurs les Juges, les deux
dispositions autour desquelles se résume le débat des rapports entre ces deux piliers du droit
international le plus profondément enraciné dans la pratique ancestrale des Etats. Il s’agit de
l’article 60 de la convention de Vienne, pris ici, répétons-le, comme expression de la coutume
internationale, et de l’article 33 du projet de la CDI sur la responsabilité. Que disent-ils l’un et
l’autre ?
Ar6ticle (convention de Vienne) :
“Une violations substantielle d’un traité bilatéral par l’une des Parties autorise
l’autre Partie à invoquer la violation comme motif pour mettre fin au traité ou
suspendre son application.”
Il s’agit là, vous en conviendrez, d’une règle d’habilitation. Quel est son bénéficiaire ? C’est
l’Etat victime de la violation du traité. Je dis bien l’Etat victime.
Voy l'ons article 33 (projet de la CDI) maintenant : “L’état de nécessité ne peut pas être
invoqué par un Etat comme une cause d’exclusion de l’illicéité d’un fait de cet Etat non conforme à
l’une de ses obligations internationales”, sauf bien sûr, les conditions particulières que nous avons
déjà étudiées dans la première partie. Cela, ce n’est plus une règle d’habilitation. C’est une règle
d’interdiction, assortie d’exceptions conditionnelles. Quel en est le destinataire ? Par définition, et
comme le dit clairement le texte, ce n’est plus du tout, comme à l’article 60, l’Etat lésé; c’est au
contraire l’auteur du fait illicite, l’Etat à priori responsable.
258
RS, p. 88, par. 4.21. -79-
19. Ainsi, dans l’un et l’autre cas, ce ne sont tout simplement pas les mêmes Etats qui sont
concernés. Le droit des traités autorise la victime à suspendre ou terminer le traité. Le droit de la
responsabilité excuse l’Etat responsable, à certaines conditions. Ils n’ont tout simplement ni le
même destinataire, ni le même objet, ni par voie de conséquence, la même fonction.
La question n’est donc aucunement de savoir si le droit de la responsabilité ajoute des
clauses de suspension ou de terminaison à celles prévues par le droit des traités, puisqu’ils ne se
situent pas sur le même plan. C’est pourquoi, comme le disait Paul Reuter, la convention de
Vienne “s’est constamment efforcée d’exclure et de réserver les conditions de responsabilité”.
Mr. President, I still have about 10 minutes. Should I go on ?
The PRESIDENT: Please proceed.
M. DUPUY : Thank you very much. Je disais donc, c'est bien pourquoi, comme le notait
Paul Reuter, la convention de Vienne “s’est constamment efforcée d’exclure et de réserver les
conditions de responsabilité”. Il précisait qu’en dépit des “contacts” et des “rapports” entre les
deux matières, “il s’agit de deux systèmes de règles qui sont séparés par une différence
259
essentielle ”, celle, précisément, qui tient à leur différence formelle.
S'agit-il là d'une logique à la française, ou formelle pour être vraie? Trompeuse clarté de la
pensée cartésienne ?
“A Court cannot remedy a breach of a treaty by reading into the treaty a
sanction or remedy for which it does not provide. This does not affect the
responsibility of the defaulting State, but the remedy will consist in the application of
the ordinary rules of international responsibility.”
On aura reconnu sir Gerald Fitzmaurice dans son Law and Procedure of the International Court of
260
Justice .
25Introduction au droit des traités, p. 153 et 158.
260
Grotius, 1985, p. 50. -80-
On aurait d’ailleurs tout autant pu citer dans ce contexte la proposition de sir Humphrey Waldock
dans laquelle la Slovaquie croyait pouvoir trouver quelque secours dans sa réplique, lorsqu'il
proposait dans son rapport de 1964 l’adjonction d’un paragraphe au projet relatif au droit des traités
déclarant simplement, pour reprendre les termes dans lesquels cette proposition est évoquée par la
réplique slovaque :
“that the failure of a State to comply with its obligations in good faith engages its
responsibility unless this failure is excusable under the general rules of State
261
responsibility” .
20. Ce disant, ces auteurs ne faisaient que reprendre le dictum de la Cour internationale de
justice dans l’affaire relative à l’Interprétation des traités de paix, lorsqu’elle notait elle-même: «il
est clair que le refus de s’acquitter d’une obligation conventionnelle est de nature à engager la
262
responsabilité internationale» . C’est alors dans ce cadre qu’interviendra la question de savoir si
entrent en ligne de compte des circonstances excluant l’illicite de l’Etat incriminé, comme c’est
précisément ici le cas pour les raisons que nous avons exposées plus haut et qui seront reprises
demain.
Personne, sauf, hier, la Nouvelle-Zélande, et, aujourd’hui, la Slovaquie, ne demande à ces
deux branches du droit d’apporter réponse aux mêmes questions ! Le droit de la responsabilité ne
dit pas cela, parce que ce n’est pas son métier. Il dit à quelles conditions restrictives un Etat qui a
suspendu l’exécution d’un traité ou l’a terminé n’encourt pas pour autant l’engagement de sa
responsabilité.
21. Pour autant, à moins de sombrer dangereusement dans l’esprit de système, il ne s’agit pas
de prétendre qu’il n’existe aucun point de contact entre le droit des traités et le droit de la
responsabilité. Ce serait évidemment absurde, puisque le second traite d’obligations établies en
application du premier. Il existe plusieurs «points de rencontre» entre ces deux droits, pour parler
comme Paul Reuter à nouveau. Le concept de «force majeure», par exemple constituait à juste titre
26Yearbook of the ILC, 1964, vol. lII, p. 7, cité par la réplique slovaque p. 91, par. 4.29.
262
C.I.J. Recueil 1950, p. 228. -81-
263
pour lui l’un de ces points de contact entre les deux domaines . Le même auteur constatait que la
survenance d’une situation rendant l’exécution impossible constitue, à la fois, un motif de mettre
fin ou de suspendre le lien conventionnel et comme une circonstance excluant l’illicite. Le droit des
traités déterminera dans quelle mesure ce fait affecte la survie du lien conventionnel. Le droit de la
responsabilité, quant à lui, définira s’il existe une circonstance susceptible d’exonérer de sa
responsabilité l’Etat auteur du fait illicite.
On comprend, dès lors, le commentaire que la Commission du droit international faisait à
l’article 61 de la convention de Vienne lorsqu’elle déclarait à son sujet qu’elle n’entendait pas
traiter «du cas général de force majeure, qui relève du droit de la responsabilité...» «D’ailleurs,»
ajoutait-elle, «l’article 73 ... réserve toutes les questions relatives à la responsabilité
264
internationale» . Ce qu’elle disait de la force majeure, on peut, bien entendu, le dire également de
l’état de nécessité.
Par conséquent, et j'en termine ainsi, il résulte des explications qui précèdent deux
conséquences concrètes à tirer pour le présent cas :
- en premier lieu, en fait, la Hongrie se trouvait bel et bien, dès 1989, face à une situation
affectant ses intérêts essentiels, l’autorisant à invoquer l’état de nécessité;
- en second lieu, en droit, rien n’interdit à la Hongrie d’invoquer l’état de nécessité dans un
contexte mettant en cause l’application de ses obligations conventionnelles.
Monsieur le Président, Messieurs les juges, je vous remercie de votre attention.
The PRESIDENT: Thank you, Professor Dupuy. The Court will now rise and resume
tomorrow morning at 10 o'clock.
The Court rose at 1.05 p.m.
26Annuaire de la CDI 1980, vol. I, p. 3, par. 8.
26Annuaire de la CDI 1980, vol. II , 2 partie, p. 78, par. 1. -82-
__________
Audience publique tenue le mardi 4 mars 1997, à 10 heures, au Palais de la Paix, sous la présidence de M. Schwebel, président